Finite Mathematics For Business and The Social and Life Sciences

Finite Mathematics
for Business and the Social and Life Sciences [
A Problem-Solving Approach ! *
FREE INSTRUCTOR'S COPY FREE INSTRUCTOR'S COPY J

Index of Applications
With Section Numbers
Applications are keyed to Chapter and Section. An item keyed to 4.2, for ex-
ample, means that it will be found in Chapter 4, Section 2. An "R" refers to
review exercises.
BUSINESS AND ECONOMICS Equity, 4.4, 4.R

Estate, 1.4. l.R
Advertising, 3.4, 3.5, 3.R, 5.4, 6.3, 6.5,
Executive board, 5.1
6.8,6.9,6.10
Exports, 7.1
Amortization, 4.4, 4.R
Assembly line, 3.3 Fabrication, 3.2
Assignment, 3.1, 3.3 Forecasting, 5.4
Bidding, 6.4 Graph theory, 2.R

Bonds, 2.5
Input-output analysis, 2.7
Break-even analysis, 2.3, 2.4, 2.6
Installment purchases, 4.4, 4.5
Budgets, l.R
Interest, 4.1
Business decisions, 5.R
Inventory, 2.1
Cash premium, 4.5 Investments, 1.1, 1.2, 1.4, 1.6, l.R, 2.1,
Cash value, 4.1, 4.4, 4.R 2.2,2.4, 2.5, 2.6, 3.1, 3.3,3.4, 3.5,
Closed Leontief system, 2.7 4.1, 4.3,4.4, 5.5, 6.9, 6.10
Commission, 1.1, 2.3, 5.2
Junk bonds, 4.3
Competing businesses, 6.7, 6.8
Compound amount, 4.1, 4.2, 4.5, 4.R Labor unions, 5.2
Compound interest, 4.1, 4.2, 4.R Liability payments, 6.4
Consumer's choices, 6.R Loans, 4.1, 4.R
Construction, 6.4
Machine failure, 6.4
Continuous compounding, 4.2
Management, 5.1, 5.2, 5.3
Contracting, 6.4
Manufacturing, 1.1, 2.5, 2.6, 3.1, 3.3,
Cost analysis, 1.4, 2.2
3.4, 3.5, 7.5
Cost and revenue functions, 1.5
Manufacturing costs, 2.2
Credit, 6.2
Marginal cost, 1.6
Debt, 4.4, 4.5 Marginal profit, 1.6
Decision making, 6.2 Marginal revenue, 1.6
Deferred payments, 4.5 Marketing, 2.1, 3.R, 5.4, 6.2, 6.5
Demand functions, 1.1, 2.5 Marketing costs, 2.2
Demographics, 6.5 Marketing survey, 5.4
Depreciation, 1.2, 1.4, 1.6 Markov processes, 2.1, 2.2, 6.5, 6.6
Discount, 1.2 Mark-up, 1.4, l.R
Mining, 3.5
Effective rate, 4.2, 4.R
Mixture problems, 1.4, 2.3, 2.4, 3.1, 3.3,
Efficiency, 7.5
3.5
Employee identification, 5.3
Non-interest bearing note, 4.R
Employment risks, 6.1
Endowment, 4.R Occupational studies, 6.5, 7.1
Equations of value, 4.5 Open Leontief system, 2.7
Organization, 5.3 Ages, 7.1
Organizational management, 5.3 Ages of new mothers, 7.1
Aging, 3.1
Payment of debt, 4.5
Agriculture, 3.1, 3.3. 3.5, 6.9, 6.10
Payoff decisions, 6.R
Agronomy, 3.5
Payroll, 5.2
Animal nutrition, 3.3
Periodic payments, 4.4
Anthropology, 4.R
Perpetuity, 4.5
Aptitude tests, 7.1, 7.4, 7.5
Price-earnings ratio, 7.2, 7.3
Automobiles, 7.1
Product analysis, 6.6, 7.R
Production, 1.1, 2.4, 3.5, 3.6, 6.2, 6.3, Cancer, 6.2
7.1, 7.5 Colds, 7.1
Profit, 1.1, 1.2, 1.4, 1.6, 2.4, 3.3 College (education fund), 4.3
Profit and cost functions, 1.6 College (expenditures), 7.1
Purchasing, 5.3, 6.9, 6.10 College (grades), 7.1, 7.2, 7.3, 7.4
College (graduates), 7.1
Quality assessment, 7.3
College (majors), 7.1
Quality control, 5.1, 5.2, 5.3, 5.5, 5.R,
College (student absences), 7.1
6.1, 6.2, 6.3,6.4, 6.R, 7.4
College (student profile), 5.4, 5.5, 7.1,
Refining, 3.3, 3.5 7.5
Resource allocation, 3.5, 3.7 Communications, 2.1, 2.2, 6.5
Rent, 4.5 Demographics, 2.1, 2.2
Retirement account, 4.3 Diabetes, 6.2, 6.6, 6.R
Revenue equation, 1.1
Diets, 1.1, 2.4, 2.5, 2.6, 3.1. 3.3, 3.6,
Revenue functions, 1.6
3.R, 7.4, 7.5
Salaries, 6.2, 7.2, 7.3 Dispensing drugs, 5.3
Sales, 1.6, 2.4, 5.3, 6.3, 6.4, 7.1, 7.4, Divorce, 5.4, 6.3
7.6, 7.R Drug analysis, 5.1, 6.2
Sales commission, 1.1 Drug effectiveness, 6.3, 7.R
Savings, 4.1, 4.3, 4.R Drug response, 6.3
Selling price, 4.R Drug (side effects), 6.3
Service calls, 5.2 Education, 3.1, 3.3
Simple interest, 1.5, 4.1, 4.5, 4.R Engagements, 7.1
Sinking fund, 4.3, 4.R Experimental design, 5.3
Stock dividend, 4.5
Stocks, 2.2, 7.1
Family characteristics, 5.4
Family planning, 5.2, 5.3, 6.2, 6.3
Stock trends, 6.2
Fertility drug, 6.3
Storage, 2.1
Fish population, 4.1, 4.2, 4.3
Supervision, 6.R
Flu epidemic, 6.R
Supply and demand, 1.2, 1.3, 1.4, 1.5
Time analysis, 3.3

Genes, 5.5, 6.1
Genetics, 2.1, 2.2, 5.5, 6.5, 6.6
Training programs, 6.3, 6.4
Transportation, 2.7, 3.3, 3.6, 5.2, 5.3
Grades, 7.1, 7.4
Graduation, 7.1
Trust fund, 4.5
Unemployment, 5.2
Heights, 7.4
Unions, 5.2, 5.5, 5.R, 6.1, 6.2
Herbicides, 2.6
Inflation, 4.1, 4.3

SOCIAL AND LIFE SCIENCES I.Q.'s, 7.R
Accidents, l.R, 5.2, 6.3, 6.4 Lottery, 6.4

Achievement tests, 7.4 Lie detection, 6.2
Continued inside back cover
Digitized by the Internet Archive
in 2010
http://www.archive.org/details/finitemathematicOOwhee
Finite Mathematics
forBusiness and the Social

and Life Sciences
A Problem-Solving Approach
Ruric E. Wheeler Samford University
Please note this annotated Instructor's Review Copy has been prepared for
review purposes only. We plan to subject this edition to another rigorous era
check to insure the accuracy of the student edition. Final bound books of the
Student Edition will be available for adoption May I. 1991. To order this text
through your bookstore, use the following ISBN: 0-03-046939-2. For further
information on this title, contact your local Saunders representative or the
Saunders Regional Office nearest you. Eastern Regional Office:
1-800-551-BOOK; Western Regional Office: -800-TH E-TEXT.
1
Saunders College Publishing

Philadelphia Fort Worth Chicago San Francisco
Montreal Toronto London
Sydney Tokyo
Copyright 1991 by Saunders College Publishing, a division of Holt, Rinehart and Winston,
Inc.
All rights reserved. No part of this publication may be reproduced or transmitted in any
form or by any means, electronic or mechanical, including photocopy, recording, or any
information storage and retrieval system, without permission in writing from the publisher.
Requests for permission to make copies of any part of the work should be mailed to Copy-
rights and Permissions Department, Holt, Rinehart and Winston, Inc., Orlando, Florida
32887.
Text Typeface: Times Roman

Compositor: Bi-Comp
Acquisitions Editor: Robert Stern
Developmental Editor: Richard Koreto
Managing Editor: Carol Field
Project Editor: Mary Patton
Copy Editor: Linda Davoli
Manager of Art and Design: Carol Bleistine
Art Director: Christine Schueler
Art and Design Coordinator: Doris Bruey
Text Designer: Arlene Putterman
Cover Designer: Lawrence R. Didona
Text Artwork: Grafacon
Director of EDP: Tim Frelick
Production Manager: Charlene Squibb
Marketing Manager: Denise Watrobsky
The cover illustration is a computer-generated image of a recently discovered surface of constant mean
curvature. This is a surface that locally minimizes area, subject to a volume constraint. The best-known
example is was the only known example. The new
the round sphere, and for a very long time the sphere
was discovered by H. Wente; further discoveries were made by
surface, one of an infinite family of tori,
U. Abresch and others. The example seen on the cover, with four lobes, is cut in half like an orange so
that one can see the inside. The numerics, using the Abresch representation, and computer graphics were
prepared by Y. Fang and J. Hoffman, respectively, at the Center for Geometry, Analysis, Numerics and
Graphics. University of Massachusetts, Amherst.
Printed in the United States of America
FINITEMATHEMATICS FOR BUSINESS AND THE SOCIAL AND LIFE SCIENCES:

A PROBLEM-SOLVING APPROACH
ISBN 0-03-046939-2
Library of Congress Catalog Card Number: 90-053334
1234 039 987654321
THIS BOOK IS PRINTED ON ACID-FREE, RECYCLED PAPER

To Dr. William D. Peoples,
who worked with me on material
that later culminated in the
publication of this book

Preface
This book provides the noncalculus mathematics needed by students in man-

agement, social sciences, and life sciences. Our primary goal is to publish a
book mathematics of the better publications in this field,
that contains the
and at the same time easily read and comprehended by the average stu-
is
dent. In addition to being readable, how does this text differ from the typical
book serving these students? The answer can be summarized in four main
points.
Problem Solving. The ability to analyze problems and translate them into
mathematical language is an important and, for the average student, a diffi-
cult skill. This book begins with a section on problem solving and encour-
ages the student to use problem-solving procedures throughout. Thought
processes and algorithmic procedures are introduced to improve problem
solving.
Future Usability. Often students ask, "Why should I study mathematics?"
One goal of this book is to answer this question for students who are major-
ing in business, economics, the social sciences, or the life sciences. We
answer the question in three ways. First, every section of this book contains
numerous applications, classified as "Business and Economics" or "Social
and Life Sciences." Second, the last section of each chapter. "Extended
Application," contains either a case study using the material of the chapter
or an example from a textbook in business, economics, or the social and life
sciences using the chapter material. (Knowing material will be covered in a
later course encourages students to study and learn.) Finally, CPA, CMA,
Preface
and actuarial exam questions are scattered throughout the exercise sets to
indicate that knowledge of the material is important in professional growth.
These are denoted by the word "Exam."
Ample Review. The only prerequisite assumed for this course is high school
algebra. However, since algebra is easily forgotten due to lack of use, the
first chapter contains ample review of basic topics of algebra. For classes
that do not need this review, we suggest studying Sections 1 and 4 of Chap-
ter 1 and then Chapter 2.
Student-Based Approach. First and foremost, this book is student oriented.
A distinct effort is made to base each new concept on the student's prior
experience or prior knowledge from the textbook. Thus, the book appeals to
intuition rather than to a formal approach.
Organization
Chapter 1 (Problem Solving and Linear Models) is the foundation chapter

covering such topics as real numbers, set theory (important in probability),
rectangular coordinates, and linear equations and inequalities (important in
linear programming). Section 4 contains the beginning discussion of prob-
lem-solving techniques. A study of lines is introduced as a foundation topic
for those who will study calculus. Various application problems are intro-
duced as linear models. Parts of this chapter may be familiar to some stu-
dents, and the instructor may find a quick review of this material to be
adequate.
Chapter 2 (Linear Algebra) introduces matrices, the properties of ma-
trices, operations involving matrices, the solution of linear systems, and
many applications. A study of the inverse of a matrix leads to an important
application called input-output analysis. Matrices are used in such topics as
game theory, linear programming, and Markov processes.
Chapter 3 (Introduction to Linear Programming) begins with an ele-
mentary introduction of linear programming from a geometric point of view.
Special attention is given to the transition from a geometric procedure to a
matrix procedure (called the simplex procedure) for solving linear program-
ming problems. Both the solution for the dual problem and the solution of
nonstandard linear programming problems are considered in this chapter.
Chapter 4 (Mathematics of Finance) presents a calculator approach to
solving mathematics of finance problems, although tables are provided in
Appendix B for those who wish to use them. All examples are illustrated
with time diagrams. A discussion of equations of value and perpetuities in
the last section of the chapter ties together the various chapter topics.
Chapter 5 (Counting Techniques and Probability) discusses probability
for finite sample spaces. Permutations, combinations, tree diagrams, and
other counting procedures are introduced to assist in solving probability
problems. Compound and conditional probabilities are introduced in order
to understand stochastic processes in Chapter 6.
Preface
Chapter 6 (Additional Topics of Probability) discusses stochastic pro-

cesses along with Bayes' formula, binomial probability, and expectation.
The remainder of the chapter deals with important applications of probabil-
ity: Markov processes and game theory.
Chapter 7 (Statistics) introduces the basic concepts of descriptive sta-

tistics.The normal distribution is used to approximate the binomial distribu-
tion and to find confidence limits for a population mean. These concepts give
the student some appreciation for inferential statistics.
Pedagogical System
A consistent pedagogical system is used in every chapter.
OVERVIEWS Each section begins with an overview, which quickly summarizes the sec-
tion's goals and relates the section to previous study or experience. If the
section's objectives are not covered in the discussion, they are listed at the
end of the overview.
STUDENT AIDS Thought boxes are provided to assist students through the critical stages of a
problem. For example:
Page 97:
Need a
here .4 -3 -1 Use -4R, + fc
Page 219:
6000 = P(l + 0.02)

1
A = $6000
= r- =
r .08
=
l -r .02
k 4
Steps in mathematical thinking are often repeated in nonmathematical terms

to ensure complete understanding.
PRACTICE PROBLEMS Simple problems are presented prior to the exercise set to check for under-
standing. Often the steps leading to a solution are given to provide additional
information before the student attempts the exercise set.
NOTES Throughout the text you will find "NOTE" in the margin to indicate an
unusual thought or to call attention to possible student errors.
CALCULATOR NOTES Also found throughout the text are suggestions for using a calculator. Often
the steps for obtaining the answer using a calculator are enumerated.
Preface
VISUAL AIDS Someone has worth a thousand words. For this reason the
said a picture is
student will find an unusually largenumber of diagrams to assist in under-

standing, such as time diagrams for annuities (see page 232).
END-OF-CHAPTER At the end of each chapter is a short test on the chapter material to assist the
TESTS student in evaluating comprehension. All the answers to these problems are
given in the back of the book.
EXERCISES Exercise sets can be either a strength or a weakness of a mathematics text-

book. The exercise sets of this book, containing more than 2700 problems,
are an important feature. The problems are arranged according to the level of
difficulty. They usually occur in matching pairs, with the answers to the odd-
numbered problems provided in the back of the book. Application problems
are presented at the end of each exercise set.
EXAMPLES The textual material of this book is augmented by more than 300 solved
examples. Every new idea is illustrated with an example.
Calculator Approach
An inexpensive scientific calculator or financial calculator should be pur-

x
chased for this course. The capability of the calculator to compute y is
important. The ability to group terms with parentheses is desirable. In addi-

tion, hoped that at least one scientific calculator with graphing capabili-
it is
ties is available to each class of students using this book. Undoubtedly,

within the next few years, many advances in calculator capabilities will take
place. We have done our best to place this book on the "cutting edge" of
calculator technology.
Supplements
Finite Mathematics for Business and the Social and Life Sciences comes
with a complete set of ancillaries that will provide invaluable help to students
and instructors.
Test Bank and Computerized Test Bank by Edmond Dixon of Tennessee

Technological University. The test bank contains more than 1500 ques-
tions for the instructor to use in tests or for homework assignments. It
is available in a printed version or in a computerized version on an IBM
disk that allows the intructor to custom-design tests.
Instructor's Manual by Henry Smith of the University of New Orleans. A
solutionprovided for every exercise in the text.
is
Student Solutions Manual and Study Guide, also by Henry Smith. This sup-
Preface
plement for the student contains the answer to every other odd-num-
bered exercise. In Professor Smith's unique system, new concepts are
keyed to answers to exercises in the text.
Prepared Tests by Leslie Cobar of the Univesity of New Orleans. Professor
Cobar has prepared five tests for each chapter and two final exams
more than 1500 questions in all.
MathPath Tutorial Software by George W. Bergeman is available for Finite
Mathematics on an IBM disk. Professor Bergeman has written new
modules unique to this text. The software is keyed to each chapter.
Acknowledgments
The following reviewers provided valuable suggestions for improving the

text:
James Arnold, The University of Wisconsin at Milwaukee

George Articolo, Rutgers University
Ronald Bensema, Joliet Junior College
Mitzi Chaffer, Central Michigan University
Sam Councilman, California State University at Long Beach
B. Jan Davis, University of Southern Mississippi
Edmond Dixon, Tennessee Technological University
Gerald Goff, Oklahoma State University
Madelyn T. Gould, DeKalb College
Jay Graening, University of Arkansas
Linda Halligan, Mohawk Valley Community College
Cheryl M. Hawker, Eastern Illinois University
Allen Hesse, Rochester Community College
Richard Hobbs, Mission College
Roseanne Hofmann, Montgomery County Community College
Robert T. Kurosaka, Massachusetts Bay Community College
Ronald M. Jeppson, Moorhead State University
Charles W. Johnson, Louisiana State University at Shreveport
Jim Loats, Metropolitan State College
Stanley M. Lukawecki, Clemson University
H. T. Mathews, University of Tennessee
Michael E. Mays, West Virginia University
Byron McAllister, Montana State University
Linda McDonald, University of Texas at Arlington
Jeffrey McLean, College of St. Thomas
Stephen Rodi, Austin Community College
Dan Symancyk, Anne Arundel Community College
James Verhanovitz, Delta College
Delores A. Williams, Pepperdine University
Preface
My thanks to Professors David Foreman, Mary Hudson, and Fred

Kelley for preparing the answers that are found in the back of this book, and
to student assistants Lucy Reed and Carol Pennington for typing, checking,
and correcting. The continual support and assistance of the editorial and
production staff at Saunders College Publishing are also gratefully acknowl-
edged.
Ruric E. Wheeler
Samford University
December 1990
Contents Overview
1
xiw Contents Overview
6
Contents
1 Problem Solving and Linear Models
1.1 Suggestions for Problem Solving 1
1.2 Foundations: Sets and the Real-Number

System 7
1.3 Solving Equations and Inequalities 18
1.4 Solving Problems with Linear Models 28
1.5 Expressing Ideas with Graphs 33
1.6 Slopes and Linear Equations 44
Extended Application: Budget Lines 54
Summary and Review Exercises 55
2 Linear Algebra 61
2.1 Getting Acquainted with Matrices 62

2.2 Matrix Multiplication 72
2.3 Solving Systems of Equations 82
2.4 Solving Systems of Linear Equations with Matrices 92
Contents
2.5 Systems with Three or More Variables or Three or More

Equations 103
2.6 Inverse of a Matrix and Its Use 114
2.7 Input-Output Analysis 125
Extended Application: Graph Theory and Matrices 134
3 Introduction to Linear Programming 141
3.1 Solving Systems of Inequalities Graphically 142

3.2 Finding an Optimal Solution 150
3.3 Graphical Solutions to Linear Programming Problems 159
3.4 A Graphical Introduction to the Simplex Method 170
3.5 The Simplex Method of Maximization 181
3.6 Minimization Using the Dual Problem 191
3.7 Mixed-Constraint Linear Programming 198

Extended Application: Linear Programming A Continuing
Success Story 206
^4 Mathematics of Finance 212
4.1 Some Comparisons of Interest Rates 213

4.2 Effective Rates, Continuous Compounding, and Geometric
Progressions 223
4.3 Amount of an Annuity: Sinking Funds 231
4.4 Present Value of an Annuity and Amortization 238
4.5 Equations of Value and Perpetuities 244
Extended Application: Decisions and the Cost of
Money 250
Contents xvii
^5 Counting Techniques and Probability 256
5.1 The Language of Probability 257

5.2 Empirical Probability and the Fundamental Principle of
Counting 266
5.3 Counting Techniques Using Permutations and
Combinations 276
5.4 Probability for the Union of Events 285
5.5 Conditional Probability and Independent Events 297
Extended Application: Decision-Tree Analysis 306
6 Additional Topics of Probability 313
6.1 Finite Stochastic Processes 313

6.2 Bayes' Formula 321
6.3 Bernoulli Trials and the Binomial Distribution 328
6.4 Expected Value and Decision Making 337
6.5 Markov Chains 346
6.6 Absorbing Markov Chains 359
6.7 Introduction to Game Theory 369
6.8 Solving 2x2 Matrix Games 378
6.9 A Technique for Solving n x 2 Games 388
6.10 Solving Games by Linear Programming 393
Extended Application: Payoff Tables and Decisions 400
7 Statistics 407
7.1 Frequency Distributions and Graphical Representations 408

7.2 Measures of Central Tendency 422
7.3 Measures of Variation 431
7.4 The Normal Distribution 441
7.5 Estimating Using Sample Measurements 451
Extended Application: A Comparison of the Binomial and Nor-
mal Distributions 457
xviii Contents
Problem Solving and
Linear Models
Experience has shown that a mastery of this chapter is fundamental for good
work in this course and for the application of mathematics to business,
economics, and the social and life sciences. Generally, students learn mathe-
matical manipulations more easily than the application of mathematical con-
cepts. However, unless you can apply the mathematical techniques of this
book to meet real needs in business, economics, or the social and life sci-
ences, this course will not have fulfilled its purpose. To accomplish this goal,
we will study the forming of models in this chapter. To assist in forming
models to represent real situations, we will introduce a number of problem-
solving techniques that will assist in converting verbal statements into math-
ematical shorthand or terminology. In this chapter we will emphasize linear
models, and a picture of the linear mathematical model will be shown as a
graph on a coordinate system. Master these concepts as you begin your
study of this book.
1.1 Suggestions for Problem Solving
Overview Most college freshmen enter their first mathematics class with more expertise in mathematical
computation than in problem solving. In this section, you will be introduced to some problem-
solving suggestions that have been helpful to others. A few will be illustrated by "trick" prob-
lems; yet, these suggestions can be used in a multitude of applications. You will study
Chapter 1 Problem Solving and Linear Models
Four steps in problem solving

How understand a problem
to
How language
to translate to a mathematical
Equations and inequalities
In the 1950s, a very successful research mathematician named George

Polya wrote a series of insightful articles and books on problem solving. He
identified four steps that have characterized problem solving from the time
of the ancient Greeks to the present day.
Four Steps in Problem 1. Understand the problem.

Solving 2. Devise a plan.
3. Carry out the plan.
4. Look back; see if your solution makes sense.
In trying to "Understand the problem," we ordinarily ask a number of

questions. Typically they include:
How many?
How much?
What was?
Who is?
Who has?
What is to be found?
EXAMPLE 1 On their way back to the university, Joy, Beth, and Dill took turns driving.
Joy drove 50 miles more than Beth; Beth drove twice as far as Dill. Dill only
drove 10 miles. How many miles is the trip back to the university? (List
some questions you would ask in order to understand the problem.)
SOLUTION Understanding the problem: Do you know how far anyone drove?
(Yes, Dill drove 10 miles.) How many more miles did Joy drive than Beth?
(50).What is the relationship between the number of miles driven by Beth
and the number driven by Dill? (Beth drove twice as far as Dill.) What are we
trying to find? (The distance back to the university. )
Although asking questions similar to the ones in the preceding example
an understanding of a problem, some additional suggestions
will usually give
may be helpful.
Understanding the 1. Read and reread the problem; look up words you don't know.
Problem 2. Identify what you are trying to find.
3. Strip the problem of irrelevant details.
4. Don't impose conditions that do not exist.

Suggestions for Problem Solving 3
Once you have carefully summarized the information given in the prob-
lem, the suggestion "Identify what you are trying to find" enables you to
find the answer easily for simple problems such as the following.
1. A farmer has 17 calves. All but 9 die. How many does he have left?
2. Refinery I produces 200 barrels of high-grade oil a day. Refinery II pro-
duces twice as much per day as refinery I. How much do both produce in
.v days? (Exercise Set 1.1, Exercise 13)
Let's consider the first of these problems (a trick problem).
EXAMPLE 2 A farmer has 17 calves. All but 9 die. How many does he have left?
SOLUTION Did you get 8? Be careful! What is the unknown? How many
calves are living? What are the data? Originally, the farmer had 17 calves
(irrelevant!). All but 9 died (important!). Hence, 9 are living>
The need to strip the problem of irrelevant details is illustrated by these

examples.
1. Dan buys only blue and brown socks. He keeps all his socks in one
drawer, and in that drawer he has 8 blue socks and 6 brown socks. If he
reaches into the drawer without looking, what is the smallest number of
socks he must take out to be sure of getting 2 of the same color?
2. Vehicles of the Acme Trucking Company average 10 miles per gallon of
fuel in the city and 15 miles per gallon in the country. A truck with fuel
capacity of 50 gallons registers 8 gallons. How much fuel must be pur-
chased to fill the tank for a 3000-mile trip? (Exercise Set 1.1, Exercise 4)
Let's look at the first problem. Students often mistakenly concentrate

on the numbers of blue socks and brown socks when trying to solve the
problem. These numbers are not pertinent. To obtain the answer, we ana-
lyze what happens as Dan draws socks one by one from the drawer.
Dan has drawn 2 socks; they are either the same color or different
colors. If they are the same color, he has a pair in 2 draws. If they are
different colors, the next sock drawn must match one of them. Thus, at most
3 draws are necessary to get a pair of socks that match.
Polya's second step, "Devise a plan," often requires a considerable
amount of creativity. Most of our plans will involve forming mathematical
models. We develop our skills for forming such models by expressing verbal
statements in mathematical shorthand such as the following.
EXAMPLE 3 The daily cost of producing lawn mowers at theApex Corporation is a fixed
cost of $5000 plus a variable cost of $100 times the number of mowers
produced that day. What is the cost of producing m mowers?
SOLUTION The variable cost is $100 times m mowers, or 100m. But this must
be added to a fixed cost of $5000. The cost C = $5000 + $100m>
EXAMPLE 4 Translate the following into mathematical language:
(a) Six added to the product of 9 and a certain number

(b) Eight taken from the product of 3 and a number
(c) A rental company uses the following expression for the first yearly de-
preciation of automobiles: Add $3000 to 10% of the initial cost.
SOLUTION
(a) 9^ + 6
(b) 3x - 8
(c) 0.10C + $3000^
Sentences that can be identified as either true or false are called state-
ments. Each of the following is a statement:
(a) 8 + 2 = 2 + 8 (a true statement of equality)

(b) 7 - 3 = 8 - 4 (a true statement of equality)
Statements of equality are called equations. Statements of inequality may be

written using any of the following symbols:
+ is not equal to
> is greater than
< is less than
Precise definitions for the terms "less than" and "greater than" will be
given later in this chapter. Your intuitive understanding of these symbols
will be sufficient for this section.
EXAMPLE 5 (a) 6 - 4 < 5 - 2 (a true statement of inequality)

(b) 5 + 4 > 2 + 1 (a true statement of inequality)
(c) 7 + 3 < 2 + 4 (a false statement of inequality)!"'
Many sentences involving mathematical variables, or unknowns, can-

not be classified as either true or false. These are not statements but rather
are called open sentences. Two examples of open sentences are equations
such as
x + 4 = 1
and inequalities such as
x + 1 < 10
A first step in forming models is writing open sentences that represent

verbal expressions.
EXAMPLE 6 Translate the following expressions into mathematical sentences:
(a) A number added to 15 is equal to 23.

(b) A number minus 3 is greater than 7.
Suggestions for Problem Solving 5
SOLUTION
(a) 15 + x = 23
(b) z - 3 > >
Later in this chapter we complete Polya's third step, "Carry out the
plan," by solving the linear models we have created.
Polya's last step, "Look back," is a must, not only for problems in this
book, but for those in life as well. In this step you determine if your answer
makes sense relative to what is given. Mathematically speaking, we call this
checking to see if our solution satisfies given conditions.
Practice Problems > 1. A mining company owns two mines. Big Knob produces 40 tons of ore a
day; Little Creek produces 20 tons. Big Knob operated x days last week
and Little Creek operated y days. Write an expression for the amount of
company produced last week.
ore the
2. Roebuck Power Company charges $5 per month for service plus $. 10 for
each kilowatt-hour used. Mike Smith's electric bill was $30. Write an
equation to find how many kilowatt-hours of electricity he used.
Answers !o practice problems are found on page 7.
Exercise Set 1.1 In Exercises 1-4, make certain to gallons. Regular gasoline sells
strip the problem of irrelevant de- for $1.08 a gallon. What is the
tails. cost of a full tank of gasoline?
1. To select the best two-person 4. Vehicles of the Acme Trucking

debate team in the United Company average 10 miles per
States, 117 teams are selected, gallon of fuel in the city and 15
with the understanding that miles per gallon in the country.
when a team loses a match, it is A truck (with fuel capacity of 50
dropped from the tournament. gallons) contains 8 gallons of
How many matches must be fuel. How much fuel must be
played to select a winner? purchased to fill the tank for a
3000-mile trip?
2. Ron buying old comic books
is
for 80, old buttons for 40, and

Express as mathematical sentences.
old toys for 600 each. He sells
each of them for 50 profit. How 5. A number added to 7 equals 10.
much does he make if he buys
6. Three times a number equals 21.
and sells 4 comic books, 4 but-
tons, and 2 toys? 7. Five subtracted from a number
equals 12.
3. Louise drives a small car, get-
town
ting 28 miles per gallon in 8. One number is 6 more than an-
and 34 miles per gallon on the other, and their sum is equal to
highway. Her fuel tank holds 15 20.
6 Chapter 1 Problem Solving and Linear Models
9. Three times a number divided by 11. Six more than twice a number is
2 is equal to the sum of the num- equal to 20.

ber and 6.
12. The sum of two different num-
10. Twice the sum of a number and bers is 12. The larger number is
4 is equal to 14. twice the smaller number.
Applications
(Business and Express the following as mathemati- for revenue R in terms of x units
Economlcs) cal statements or sentences. sold. Hint: Revenue = (price per
unit) (number of units sold).
13. Production. Refinery I produces
200 barrels of high-grade oil a 18. Cost Equation. In the manufac-
day. Refinery II produces twice turing of the x units in the pre-
asmuch per day as refinery I. vious exercise the fixed cost is
How much do both produce in x $1000 plus a variable cost of

days? $2.40 per unit for labor and ma-
terials. Write an expression for
14. Manufacturing. Belcher Corpo-
cost c.
ration makes valves and reduc-
ers. A valve requires 1 hour on 19. Supply Curve. A supply curve in-
machine A and 2 hours on ma- dicates the number of units of a

chine B. A reducer requires 3 good or service that will be of-
hours on machine A and 4 hours fered for sale at different prices.
on machine B. Four reducers Write an equation for the supply
and x valves were produced last curve if price p is 20 plus the
week. How many hours was ma- product of 2 and the number of
chine A in operation? Machine units (x) for sale.
B?
20. Demand Curve. A demand curve
15. Investments. An investment club indicates the total quantities that
has $27,000 to invest in bonds of purchasers will buy at different
two types, government securi- prices. Write an equation for the
ties and junk bonds. Govern- demand curve if price/? is 20 less
ment securities yield 7%, and the product of 2 and the number
junk bonds yield 10%. If x dol- of units x people will buy.
lars are invested in government
21. Profits. At the Argo Corporation
securities, what is the total
this year's profits P will exceed
yearly interest?
last year's profits of $620,000.
16. Profit. At Pizitz, a suit which Express this mathematically.
costs $100 sells for x dollars.
= 22. Sales Commissions. Each sales-
What is the profit? Hint: Profit
- man in Wargo Sales earned
Selling Price Cost.
more than $20,000 in commis-
17. Revenue Equation. Suppose a sions Express a mathematical
c.
manufacturer sells a product for relationship between x, John

$5 per unit. Write an expression Smith's earnings, and $20,000.
Foundations: Sets and the Real-Number System
Applications (Social
and Life Sciences) 23. Diet. Food I contains 30 protein acre. What is the labor require-
and food II con-
calories per unit ment for w acres of wheat and c
tains 20. Abdul eats x units of acres of corn? What is the cost
food I and y units of food II. of fertilizer and other costs for w
How many protein calories does acres of wheat and c acres of
he get? corn?
24. Medicine. Medicine mixture I

26. Psychology. An
intelligence quo-
contains 2 milligrams per gram tient (IQ) is found by dividing

of drug X (for high blood pres- the mental age .v by the chrono-
logical age and then multiplying
sure). A gram of medicine mix-
ture contains 4 milligrams. At this result by 100. Write an ex-
II
the Morningside Hospital, g\ pression for the IQ for a 10-year-

old.
grams of I and g 2 grams of II are
administered to a patient. How 27. Population Growth. A city with a
much drug X does the patient re- population of 60,000 is increas-
ceive? Hint: Give the answer in ing at a uniform rate of 4% of
milligrams. this year's population. What will
the population be in t years?

25. Agriculture. George has a 50-acre
farm on which he plants two 28. Wildlife Growth. The growth rate
crops, wheat and corn. Wheat of fish in a lake is 10% per year
requires 2 days of labor per acre, of the current number of fish. If
and corn requires 3. Fertilizer there are 20,000 fish in the lake,
and other costs amount to $40 how many will there be in /
for wheat and $30 for corn per years?
Answers to \> 1. 40* + 20y tons of coal were produced last week.
Practice Problems 2. The equation 5 + 10.v = 30 relates the number of kilowatt-hours
. ,v to the
monthly electric bill.
1.2 Foundations: Sets and the Real-Number System
Overview You use numbers every day, but you seldom think about what kinds of numbers they are. Yet,
numbers will be important to you throughout life. In this section the real numbers (the kind you
will use in this course) will be described, and number lines will be introduced to aid your
understanding of the concept of real numbers. The four fundamental operations (addition,
subtraction, multiplication, and division) will be presented so as to help you in solving equations
throughout this book. A quick review of these operations will be presented for negative num-
bers and fractions. In addition we will review the basic concepts and terminology of set theory in
order to provide a convenient mathematical language and notation.

A Review of Set Terminology
We describe a set as follows.
Definition: A Set A set is a collection of objects or symbols possessing a property that enables one to
determine whether a given object is in the collection.
We sometimes say a set is a well-defined collection, meaning that,

given an object and a set, we are able to determine whether or not the object
is in the set. The individual objects in a set are called elements of the set. The
relationship between objects of the set and the set itself is expressed in the
form "is an element of" or "is a member of."
Definition: xEA means x is an element of set A.x (.A means x is not an element of set A.
Element of a Set
Often it is possible to specify a set by listing its members within braces.

This method of describing a set is called the tabulation method (sometimes
called the roster method). The set of counting numbers less than 10 can be
written
{1, 2, 3,4, 5,6,7, 8,9}
A remains the same regardless of the order of the elements within

set
it. For example, {1, 2, 3} is the same set as {2, 1, 3}, {3, 2, 1}, {3, 1, 2}, {1, 3,
2}, or {2, 3, 1}. In fact, two sets are said to be equal if they contain exactly the
same elements. If A = B, then A and B have exactly the same elements.
EXAMPLE 7 Glenda, Cathie, and Marcia are the only counselors in the admissions office.
They constitute the set A = {Glenda, Cathie, Marcia}. Cathie G A; Linda
A. Can you identify other elements of set Alt
Sometimes sets have so many elements that it is tedious or difficult or

even impossible to tabulate them. Sets like this may be indicated by a de-
scriptive statement or a rule. The following sets are well specified without a
tabulation of members: the counting numbers less than 10, the even numbers
less than 1000, the past presidents of the United States, and the football
teams in Pennsylvania.
Frequently, three dots (called an used to indicate the omis-
ellipsis) are
sion of terms. The set of even may be written
counting numbers less than 100
as {2, 4, 6 98}. This notation saves time in tabulating elements of large
sets, but it can be ambiguous unless the set has been specified completely by
another description. For example, {2,4,. 16} could be {2, 4, 8, 16} or it
. . ,
could be {2, 4, 6, 8, 10, 12, 14, 16}.

An ellipsis is also used to indicate that a sequence of elements con-
Foundations: Sets and the Real-Number System 9
tinues indefinitely. For example, consider the set of natural, or counting,

numbers:
{1, 2, 3,4, 5, . . .}
The set of natural numbers an example of an

infinite set, described infor-
is
mally as one that contains an unlimited number of elements. In contrast, a

finite set contains either no elements or a natural number of elements.
Definition: An Empty Set A set that contains no elements is called the empty, or null, set and is denoted by either I
or{ }.
The relationship between two sets such as A= {1, 3, 5, 7} and/? = {1,2,

3, 4, 5, 6, 7, 8} is described by the term "subset."
Definition: A Subset Set A is said to be a subset of set fl, denoted by A C 8, if and only if each element of A is
an element of S.
For example, if-P = {1,4, 7} and Q = {4, 7, 1}, then/5 C Q since each element
of P is an element of Q.
Since all dogs are animals, the set of dogs is a subset of the set of
animals. Moreover, the set of dogs is a proper subset of the set of animals,
since there are animals that are not dogs.
Definition: Set A is said to be a proper subset of set S, denoted by A c B, if and only if each element
A Proper Subset of A is an element of B and there is at least one element of 6 that is not an element of A.
In Figure 1, A is a proper subset of 5, or A C B.
FIGURE 1
If a discussion is limited to a fixed set of objects and if all elements to

be discussed are contained in this set, then this "overall" set is called the
universal set, or simply the universe, U.
A Venn diagram, such as that in Figure 2, is a very useful method for
representing the relationship between a set and the universe. The universal
set can be considered as located within region U bounded by the rectangle,

and the set under consideration as within the region bounded by the circle
(or some other closed figure within the rectangle). To write x G A, or x is an
element of set A, means that x is a point in the circular region A.
FIGURE 2
The region outside set A and inside the universe (J represents the
complement of A, denoted by A'.
Definition: The complement of set A is the set of elements in the universe that are not in set A.
Complement of a Set
If the universe consists of all bonds, and if A consists of all bonds that
have interest rates exceeding 8%, then the set of all bonds bearing interest of
8% or less is the complement of A. In Figure 3 the shaded region outside the
circle but inside the rectangle represents A'.
FIGURE 3
EXAMPLE 8 If the universe U = {a, b, c, d] and A = {b, c}, find A'.
SOLUTION
A' = {a, d) The elements that are in U but not in A

Additional topics of set theory are presented in Chapter 5>
The Real Number System

We are ready now to review some concepts of the real number system. The
numbers we use most of the time in this book are the real numbers. There are
many subsets of real numbers, as seen in Figure 4.
Real numbers
Irrational numbers
itegers
1
Fractions that
cannot be simplified
to integers
Negative Whole
integers numbers
Positive integers
(or natural
numbers)
FIGURE 4
You will use these numbers in your study of this book, often without
thinking of the name of the numbers you are using. However, numbers such
as
{1,2,3,4.. . .}
numbers or natural numbers; these numbers along with

are called counting
make up the whole numbers. The integers consist of
{ -4, -3, -2, -1,0, 1, 2, 3,4, }
Rational numbers are numbers that can be expressed as quotients of integers

such as | and -f. Some rational numbers can be simplified to be integers
(such as | = 2) and some rational numbers are expressed as fractions (such
as f). When a rational number is written as a decimal, either it is terminating,
such as 2.13 (which is fooh or it is a repeating decimal such as 2.333 . . .
(which is 2J). Irrational numbers are usually classified as numbers that are
not rational. In this book, we will work with numbers such as V3 and V7,
which are irrational, tt is an irrational number, and so is a nonrepeating
decimal number such as 3.131131 113 (Note that the l's increase by
. . . .
one each time.) Do you understand that -2 is a negative integer, is an

integer, is a rational number, and is a real number?
There is a one-to-one correspondence between the real numbers and

the points on a line (to each real number there corresponds one and only one
point on the line, and vice versa). To illustrate this concept, we draw a line
with a uniform scale of a convenient size that has the positive direction
toward the right. This line is called the real-number line, or real line; the real
number corresponding to a point on the line is called the coordinate of the
point, and the point is called the graph of the number. The point correspond-
ing to is called the origin. Points or graphs representing the irrational
number - V10 and the rational numbers 2 and 6.5 are shown in Figure 5.
sIlO
-4 h H
I I I
\ I I I-
FIGURE 5
The basic operations on the real numbers are addition, subtraction,

multiplication, and division. Certain properties of these operations are im-
portant.
With your calculator add 42 + 33. Then clear and add 33 + 42. Note
that you get the same answer, 75. You have demonstrated that the order in
which numbers are added does not affect the sum. In general, a + b = b + a
is called the commutative property of addition.
Now with your calculator multiply two numbers in one order and then
reverse the order. Did you get the same answer? You have illustrated the
commutative property of multiplication, ab = ba.
CALCULATOR NOTE With your calculator add 41 +27. Then add 33 to this sum, (41 + 27) + 33.
Now, can you find the sum of 41 + (27 + 33)? That is, add the sum of 27
to 41
and 33.
Try 4100270330
Did you get the same answer? You have illustrated that when adding more
than two numbers you may arrange the numbers in any order for addition.
In general we state the associative property of addition as a + (b + c) =
(<(!/') c. In a like manner, the associative property of multiplication is
stated as a(bc ) = (ab)c.
Now with your calculator, illustrate the distributive property of multi-
plication over addition, a(b + c) = ab + ac and (b + c)a = ba + ca. For
example, add 42 + 37. Then multiply the answer by 53, 53(42 + 37).
Now multiply 53 times 42 and 53 times 37 and add the results. Did you
get the same answer?
Zero is called the additive identity because a + = + a = a. Like-
wise, 1 is the multiplicative identity since I a = a 1 = a.
With your calculator you can show that
42 + (-42) =
These two numbers are said to be additive inverses. In general, a and -a are
additive inverses. Likewise, a and \la (where a + 0) are multiplicative
inverses since a{\la) = 1. Ma is often written as a '.
CALCULATOR NOTE Did you have trouble getting -42 into your calculator? Try 42 1
+/- |
. Do you
see a -42 on the display?
The "absolute value" of a number v, denoted by |.v|, is always positive

orO.
Definition: x if x >
Absolute Value -x if x <
This definition states that the absolute value of a nonnegative number is

that number itself, and to get the absolute value of a negative number,
change its sign (make it positive).
The operations of addition and multiplication on the set of real num-

bers are performed by using the following procedures for signed numbers.
Definition: Addition and 1. To add two signed numbers with the same sign, add their values and affix their
Multiplication of common sign.
Signed Numbers
-2 + -5= -(2 + 5)
2. To add two signed numbers with unlike signs, find the difference in their absolute
values. Affix to the answer the sign of the number whose absolute value is the largest.
-2 + 5 = +(5-2) = 3
-7 + 4 = -(7 - 4) = -3
3. The product (or quotient) of two numbers with unlike signs is negative.
-3-2 = -6and4 + -2 = -2
4. The product (or quotient) of two numbers with like signs is positive.
-4 --3 = 12 and -20 + -5 = 4
NOTE x should not be read as "negative v."' since x can be cither a positive or
negative number; that is, jr can
be positive if x is negative. For instance,
if x = -5, then -v is 5. Read x as the "additive inverse of" or "the
opposite of" x, and you'll save yourself a lot of confusion.
EXAMPLE 9 Find -(2v - 4a - 3), when v = 1.
SOLUTION -(2 1 - 4 1 - 3) = -(2 - 4 - 3) = -(-5) = 5
NOTE Always do multiplications and divisions in the order they occur in the prob-
lem, before additions and subtractions. This is the order in which most
calculators will perform the operations.
EXAMPLE 10 (a) |-3| = 3 and |17| = 17
(b) -(-6) = 6
(c) 5 + (-9) = -(9 - 5) = -4 1
9 is larger than 5 |
-3 + 5 =(5 - 3) = 2 1 5 is larger than 3 |

(d) -5 + (-6) = -(5 + 6) = -11

(e) (-3) (5) = -15

Unlike signs
(f) (-4)
(-8) = 32 1 Like signs 1
Properties of rational numbers expressed as fractions are summarized

as follows.
Properties of Fractions (a) cplcq = pi

/l\ 3 l
(b) _3 + 6 --3-(g)-- g - "2
2 5
= = _
(O-j 7 (~3/W T5
Sum-of-digits depreciation provides an excellent application of the use

of fractions. Suppose that an asset has a life of 6 years and a value of $1000.
To find the sum-of-digits depreciation, we add the natural numbers repre-
senting each of the 6 years:
1+2 + 3 + 4 + 5 + 6 = 21
We then use 21 as the denominator and each year in reverse order as the
numerator (that is, year 1 would be 6/21) to find the depreciation by multiply-
ing the resulting fraction by the initial value of the asset.
2. List within braces the members

of the sets below.
(a) The counting numbers less

than or equal to 16
(b) The set of even counting
numbers
(c) The set of women presidents
of the United States
3. Are any of the sets in Exercise 2

empty sets?
4. Which of the following sets are

well defined?
(a) The set of great baseball

players
(b) The set of beautiful horses
(c) The set of students in this
class
(d) The set of counting numbers
smaller than a million
5. Classify the following numbers

as natural numbers (N), integers
(I), rational numbers (Q), irra-
tional numbers (H), and/or real

numbers (R). (For example, -5
is I, Q, and R.)
(a) V3
(b) I
(c) 0.125
17. -8 [- - 4)
18. -11 | - 3
|
(6-5)
19.
37. Profit. A merchant sells a suit 38. Use the sum-of-digits method to
for $87, thereby gaining a profit find the yearly depreciation of
of 30% of cost. What was the an office computer valued at
cost of the suit? $5000 with a life of 8 years.
and Life Sciences) 39. Muscle Efficiency. Sometimes the muscle is contracted. Find a
algebraic expressions are used value for E when / = 0, 0.5, 4s,
to describe reactions as they oc- and 5.
cur in time. For example,

40. Politics. Lawrence Soafer is
1 - 0.2r running for city mayor. He ex-
pects to spend 60% of his cam-
paign budget on TV advertise-
has been used to describe mus- ments. If he spends $36,000 on
during maximum
cle efficiency TV advertisements, what is his
contraction over the time / that
campaign budget?
Answers to C> 1. 4(3 + 2) = 4(5) = 20. 4(3) + 4(2) = 12 8 = 20. Thus, 4(3 + 2) = 4(3)
Practice Problems + 4(2).
2. The answer is 2.
3. The answer is 0.01.
1.3 Solving Equations and Inequalities
Overview This section should be a review of solving equations and inequalities. However, a knowledge of
this material is essential in the chapter on matrix theory and the chapter on linear programming.
We consider the following topics:
Operations that give equivalent equations

The definition of a solution of a linear equation
Solutions of linear equations
Definitions of less than and greater than
Solutions of linear inequalities
The equation 2x 1 = 13 is of the first degree in x (that is, the power of

x is 1) and is called a linear equation. If the number 8 is substituted for x in
this linear equation, 2(8) - 1 equals 15, not 13. Thus, the result is false. If the
number 7 is substituted for x in this linear equation,
Check: 2(7) - 1 = 13
13 = 13
the result is true. Thus, 7 is said to satisfy (or to be a solution of) the equation
2x 1 = 13. In general, we have the following definition.
Solving Equations and Inequalities 19
Definition: Solution of an If an equation (or inequality) involves only one variable and there is a number that, if
Equation or Inequality substituted for that variable, makes the equation (or inequality) a true statement, then
that number is called a solution of the equation (or inequality).
Two equations are said to be equivalent if and only if they have the
same solutions. The following operations produce equivalent equations.
Addition and For any real numbers a, b, and c, if an equation a = b is true, then
Multiplication Properties
1. a c = b c is true.
2. ac = be is true.
3. ale = blc is true if c # 0.
EXAMPLE 13 Solve 3.v + 5 = 11.
SOLUTION To isolate the term involving x on a side by itself, we use Proper-

ties 1 and 3.
(3.v + 5) - 5 = 1 1 - 5 Prope rty 1
3.v = 6
(3.r) -3 = 6-3 Property 3
x = 2
Inaccordance with Polya's fourth suggestion, the x = 2 should be checked

by substitution.
Check: 3(2) + 5=11
6 + 5 = 11
11 = 11
The solution of 3.v + 5 = 1 1 is 2>
EXAMPLE 14 Solve ^ -3 = 1.
(i
3+3=1+3 Pro perty 1
2" 4
2@- 2- Property 2
x = 8
Check: 2- 3=1
4-3=1
1 = 1
The solution of -z 3 = 1 is 8>
EXAMPLE 15 Solve |jc + 2| = 3.
SOLUTION
= .v + 2if;t + 2>0 [Definition of absolute value

Ix + 2|
-(.v + 2) if x + 2 <
When .v + 2 = 3
.v = 1 |
Subtract 2
When -(.v + 2) = 3
-x - 2 = 3
-x = 5 |
Add 2
x = 5 Multiply by 1
Check: |-5 + 2| = |-3| = 3
|1 + 2| = |3|
= 3
The solutions for \x + 2\ = 3 are -5 and 1>
EXAMPLE 16 Solve 3x - 2(2jc - 5) = 3(.v - 2).
SOLUTION This time we need to get all terms involving x on one side of the
equation. First we simplify each side.
3* - 2(2* - 5) = 3U - 2)
3x - 4x + 10 = 3.v - 6
-x + 10 = 3.v - 6
(-3a:) + (-a) + 10 = (-3.v) + 3x - 6 [Add -3* to both sides]
-4jt + 10 = -6
-4a- = -16 Add -10 to both sides]
a- = 4 Divide both sides by -4

Check: 3 4 - 2(2 4 - 5) = 3(4 - 2)
12 - 6 = 3
2
6 = 6
The solution is 4>
The procedures for solving equations in the preceding examples may

also be used for solving equations involving letters.
EXAMPLE 17 Solve / = Prt for t.
SOLUTION
L
Pr
= 2 =
Pr
t Divide by Pr =0|
Check: I = Pr (j-)
I = I
The solution for / in / = Prt is (Pr + 0)>
EXAMPLE 18 Solve y = mx + b for x'. (m + 0)
SOLUTION
y - b = mx [Subtract b \
y-b _ Divide by m
Check: v = m \
:
+ b
y = (y - b) + b
y = y
The solution for* is x = (y - b) I m where m + 0>
The procedures used in solving the preceding equations can be used for
calculator solutions.
EXAMPLE 19 Solve (a) 3.126a: = 12.87912

(b) x + 16.221 = 74.16
SOLUTION
12.87912 , .
(a)v = 4 12
^l2^ = -
Check: 3.126(4.12) = 12.87912
12.87912 = 12.87912
The solution is 4.12.
(b) .y = 74.16 - 16.221 = 57.939
Check: 57.939 + 16.221 = 74.16
74.16 = 74.16
The solution is 57.939>
:: CALCULATOR NOTE Using the parentheses keys on a calculator, you can solve a linear equation
keeping everything within the calculator without writing down intermediate
steps. For example, we can solve
16.1 v - 14.2 = 13.3 v + 84.3
as follows:
84.3 |T] 14.2 HE '6.1 H EH 13.3
Try this on your calculator and see if you get 35 . 1 7857 43 Note the algebraic
1 .
operations performed:
(16.1 - 13.3).v = 84.3 + 14.2
_ 84.3 + 14.2
* ~ 16.1 - 13.3
Do you find these steps in the calculator solution?

Most calculators perform operations with eight- or ten-place accuracy.
Round your answer to whatever accuracy you desire. However, remember
that your check will be no more accurate than your rounded solution. Sup-
pose you use 35.179 as the solution. Let's look at the check.
Check: Use your calculator.
552.1819 552.1807
Note that the check is accurate to five digits. (You rounded your answer to
five digits.)
jjl CALCULATOR NOTE The reciprocal key is also very useful in solving equations. To solve 14.3.V =
17.1 - 8.3.v, note that
(14.3 -f 8.3)jt 17.
v = 17.
(14.3 + 8.3)
14.3 FH 8.3 RflT^r^ 17.1 R

Multiplying by the reciprocal of the number, Ix is the same as dividing by |
1
|
,
the number. Did you get 0.75664? Now check your answer.
The procedures used for solving inequalities are almost the same as
those used for solving equations. However, let us first define less than and
greater than.
Definition: Less Than If a and b are any real numbers, then a is said to be less than b, denoted by a < b, if and
only if b - a is positive; that is, if and only if there exists a positive number c such that a +
c = b.
EXAMPLE 20 -7 < -5 because -5 - (-7) = 2 (which is positive); that is, because

-7 + 2 = -5 and 2 is positive.
Definition: Greater Than If a < b, then b > a (read b is greater than a)
Sometimes the equality symbol iscombined with inequalities to state a

"less-than-or-equal-to" relationship. The statement "a < b" means "a < b
or a = b" or "a is not greater than /?."
EXAMPLE 21 (a) -4 < -4 is true because -4 is equal to -4.

(b) -3 < -2 because - 3 < -2>
The concepts of less than and greater than are easy to illustrate by
points on a number line. For example, .r < 4 is represented by all the points
on a number line to the left of 4. The graph would contain an open circle at 4.
The graph of v < 4 includes all of the preceding points and the point x = 4. In
Figure 6(a), x = 4 is indicated by a closed circle; x > 9 is represented by all
points to the right of 9; and 2 < x < 5 (this means 2 < x and for the same x,
x < 5) is represented by all points between 2 and 5 on the number line. Figure
6(b). Note that the open circles indicate that 2 and 5 are not values of .v.
4 i-
(b)
FIGURE 6
Inequalities are often written in interval notation as indicated in Table

1. Although the symbol (infinity) does not represent a real number, by
(a, c) we mean an interval starting at a and extending indefinitely to the
right. Likewise ( , a) extends indefinitely to the left.
TABLE 1
Interval
(b) If you multiply both sides of an inequality by a negative number, you

change the direction of the inequality sign.
Any real number, substituted for a variable in an inequality, that makes

the inequality a true statement is a solution. All solutions comprise the
solution set. Two inequalities that have the same solution set are said to be
equivalent. The operations listed above produce equivalent inequalities and
thus can be used to solve inequalities.
EXAMPLE 22 Find the solution set of x/3 - 2 < 4.
SOLUTION
^ < 6 I
Add 2 to both sides
x < 18 [Multiply both sides by 3 i
Consider all real numbers less than 18. By reversing our steps we discover
that these real numbers satisfy x/3 - 2 < 4.
x < 18 I Given I
"zz < 6 |
Divide both sides by 3
2 < 4 Subtract 2 from both sides
The solution set for x/3 - 2 < 4 consists of all real numbers less than 18 or*
in ( oo, 18). The graph of this answer is shown in Figure 7>
'
I I l I I l
FIGURE 7 -is -12 -6 o 6 12
EXAMPLE 23 Find the graph of the solution set of -3x + 2 < 7.
SOLUTION
-3x < 5 Add -2 to both sides
5 -1
x > Multiply both sides by
-3
The solution set consists of all real numbers greater than -f or x in (-^, *>).
See Figure 8>
FIGURE 8
'
-2-f
I
'
-
-10
\ 1
12
1 \
Often inequalities involve absolute values. Let a be any positive num-

ber. Then|jc| < a if and only if -a < x < a; likewise |jc| > a if and only if x >
a or x < -a. That is, |jc| < 6 means -6 < x < 6, and |.v| > 4 means either x <
-4 or x > 4. This concept is used to solve inequalities such as the following.
EXAMPLE 24 Graph the solution of |2 - 6jc| > 4.
SOLUTION
35. -1 ^ x + 5 < 6
18. '= + 3 < 4 - -
-< 3 * + 2>^
36
So/uf /or the variable indicated. 5
!9. A=P + p,, ;

,-
3?> 5 s
<2a+1>
< 8
20. / = A(\ - dt); t
(2jr + 5)
21. y = mx + b; b 38. -2 < . <
2
22. v = mx + b\
39. Ijc - 21 < 7
23.5 = -;
-
-; r 40. |x-5|<-l
(1 r)
41. \2x + 2| > 4

24. L = a + (n - \)d\
42. |2.v - 41 > 6
44.
1.4 Solving Problems with linear Models
Overview Students often have difficulty solving word problems. Wouldn't it be great if there were a magic
formula that would make all word problems easy? Unfortunately, such a formula doesn't exist.
There is, however, a procedure to help you formulate equations for word problems, and after a
brief introduction to models, we will practice using it. Using a calculator will also help you to
master the technique of solving word problems.
Problems that arise in business, economics, or the sciences are usually

too complicated to be denned precisely by a mathematical formula. How-
ever, we are often able to formulate a model that satisfies the parameters of a
problem fairly well.
EXAMPLE 25 Cost Analysis. and prices are analyzed over a short pe-
In business, costs
riod in terms of the manufacturing costs. (Of course, over a long period there
are many economic variables: inflation, supply, demand, etc.) Usually the
manufacturing costs consist of fixed costs (which remain constant over the
short term) and variable costs (such as labor, materials, distribution, etc.),
which depend on the number of items produced. From this discussion, write
an expression for the total cost (C) if the fixed cost is $2000 and the variable
cost is $1.20 per item produced. From this linear model, determine the cost
of producing 1000 items.
SOLUTION Let x represent the number of items to be produced. The variable

cost per itemis $1 .20. Thus, the variable cost of producing x items is $1 .20.x.
Therefore, the total cost is given by
C = 2000 + 1.20*
To find the cost of producing 1000 items we substitute x = 1000 to obtain
C = 2000 + 1.20(1000)
= $3200
EXAMPLE 26 Supply and Demand. Supply (5) and demand (>) expressed as price in dol-
lars per item for a given commodity are often given as a linear model in terms
of the number of items .v produced (or demanded). For example, consider a
demand expressed as p = (400 - 5x)l2 and a supply as p = 5x12.
(a) Which has a larger price, D or 5, when x is 10?
(b) 20?
(c) 50?
(d) 60?
(e) For what x are supply and demand equal? (The common value of supply
and demand is called the equilibrium price. Set the expression for S
equal to the expression for D and solve for x.)
Solving Problems with Linear Models 29
SOLUTION
When you complete step 3 above, you usually have a mathematical

expression to solve. This of course involves Polya's third step, "Carry out
the plan," for which we will study procedures throughout this book. Finally,
we certainly want to check to see if our answer or solution makes sense,
Polya's fourth step.
EXAMPLE 27 Susan has $10,000 invested, a portion at 8% and another at 10%. If her
annual interest is $920, how much does she have invested at each rate?
SOLUTION Guess that she has $8000 invested at 10%. Then she would have
$10,000 - $8000 = $2000 invested at 8%. Her guess
total interest for this
would be
8000 (.10) + 2000 (.08) = ?
Quickly, with a calculator, you note this is $960, which

the $920 is more than
given as interest. Now let's go back and replace our guess
$8000 by %x. Note
the wording, "Suppose she has $.v invested at 10%." Then she would have
$(10,000 - .v) invested at 8%. Her total interest would be
[jc(.IO) + (10,000 - ,v).08] = $920
You now have a linear equation representing the verbal problem. We solve
.10* + (800 - .08*) = 920 ; Subtract 800
.02.* = 120 |
Divide by .02 |
x = 120/. 02
x = $6000
Now we look back to see if our answer makes sense.
$6000 invested at 10% is $600
$10,000 - $6000 = $4000 invested at 8% is 320

Total interest: 920
Thus, our answer seems to be correct.
Did your guess and calculator approach assist in writing the equation?
Let's try this process again.
EXAMPLE 28 A will provides that an estate is to be divided among a wife and two children.
First the wife is to receive $10,000. From what remains, the wife is to receive
twice as much as each child. If the estate is valued at $50,000, how much
does each receive?
SOLUTION We begin by guessing that each child receives $5000. Do you see
that the wife receives $10,000 +2($5000)? Thus, the total received by the
Solving Problems with Linear Models 31
wife and two children is
wife child I child 2
[$10,000 + 2($5000)] + $5000 + $5000
Quickly you can see the answer of $30,000 is much less than the total estate
Now replace the guess $5000 by $.v for the amount that each child
of $50,000.
receives.
wife child 1 child 2
$10,000 + 2x + x + x
Therefore, the equation is
($10,000 + 2.v) + 2.v = $50,000 [Combine like terms
$10,000 + 4.v = $50,000 (Subtract 10,000 |
4.v = $40,000 Divide by 4
x = $10,000 (each child's share)
$10,000 + 2x = $10,000 + 2($10,000)
= $30,000 (wife's share)
The answer is the wife receives $30,000 and each child receives $10,000.
Now verify that the answer satisfies the facts. Each child receives
$10,000 and the wife, $30,000. Does the wife receive $10,000 plus double
what a child receives? The answer is yes. Does the sum of the allotments
equal $50,000? The answer is again yes because
$30,000 + $10,000 + $10,000 = $50,000
Practice Problem > 1. Mark has $20,000 invested, part of which is invested at 10% and part at
8%. If his interest is $1760 a year, how much does he have invested at
each rate? Guess the amount invested at each rate and compute the
interest to get a feel for the problem. Then set up an equation and solve
for the answer.
Answer to practice problem is found on page 33.
Exercise Set 1.4 Work Exercises I through 6 by first Tom many books as
has twice as
guessing an answer. With a calcula- Joe. Together they have 75
tor, determine whether your guess is books. How many does each
too large or too small; then substi- boy have?
tute an unknown for your guess.
A rectangle has a length 3 inches
1. Carol is twice as old as Ed. Ed is less than 4 times the width. If its
10 years younger than Carol. perimeter is 38 inches, what are
How old is Ed? the dimensions of the rectangle?
The longest side of a triangle is are $0.75 cheaper than adult

twice as long as the shortest side tickets. If total ticket receipts
and 2 inches longer than the are $1680, what is the cost of an
third side. If the perimeter of the adult ticket? A child's ticket?
triangle is 33 inches, what is the
length of each side? Exam. The Akron Company
sells ties for $6.00 each. Vari-
An automobile radiator contains
able costs are $2.00 per tie.
4 gallons of a solution that is
Fixed costs per week are
10% antifreeze and 90% water.
$37,500. How many must be
ties
How much solution must be
week
sold per to break even?
drained and replaced with pure
antifreeze to obtain a solution (a) 9375
that is 25% antifreeze?
(b) 9740
At weekend performances a
(c) 11,029
theater sells 400 adult tickets
and 520 children's tickets, which (d) 12,097
Applications
(Business and 8. Electricity. Mr. Smith's elec- decision on which coffee is the
Economics) tric billwas 6 times Mr. Jones' best buy; so he buys 1 pound of
bill. The two bills totaled $84. each brand and spends $5.08.
What was the cost of each man's How much does a pound of each
bill? brand cost?
9. Investments. A woman has an 12. Mixture Problem. A grocer

annual income of $6500 from mixes 40 pounds of $1.60-a-
two investments. She has pound nuts with 60 pounds of
$15,000 more invested at 10% $1.20-a-pound nuts. If he wants
than she has invested12%. at to receive at least the same
How much does she have in- amount of money when he
as
vested at each rate? sold the nuts separately, how
10. Investments. A sum of $2000 is much should he charge for the

invested, part at 8% and the re- mixture?
mainder at 10%.
Find the
amount invested at each rate if
13. Estate. A man leaves I of his
estate to his wife, | plus $1,000
the yearly income from the two
to each of three children, and
investments is $180.
the remainder to his pet dog. If
11. Mixture Problem. Carl is pricing he has a $100,000 estate, how
different brands of coffee. Brand much does the dog get?
A sells for $0.25 a pound more
than brand B, and brand C sells 14. Mark-up. Thomas Sporting
for $0.37 less than twice the Goods buys boots for $69.95 and
price of brand B. Carl decides to sells them for $110.50. What is
taste the coffee before making a the percentage of mark-up based
Expressing Ideas with Graphs 33
on cost? (Cost + Mark-up x the cost. What was the cost of

Cost = Selling Price) the suit?
15. Profit. Fred Moss wants to sell 17. Supply and Demand. The sup-
shirts for $14.50. If he marks up ply for a certain brand of coffee
his items 35% over wholesale is given by p = %x (where .v is the
cost, what is the greatest number of pounds of coffee and

amount he can pay for the p is the price), and the demand is
shirts'? given by p = $2.00 - |.v. By set-
ting supply equal to demand,
16. Cost. A merchant sold a suit solve for the x that gives equilib-
for $140, thereby gaining 30% of rium (supply equals demand).
and Lite Sciences) 18. leisure Time. If you spend 65% 1980. then
of your leisure time reading,
y = 1.5/ + 430
how much leisure time do you
have if you spend 13 hours a Solve for t in terms of y and then
week reading? find when the school can expect
a 450 verbal score. Will this ac-
tually happen? Why or why not?
19. Temperature. F = f C + 32
relates temperature in degrees 21. Protein. One serving of Wheats
Celsius temperature in de-
to cereal contains 4% of the recom-
grees Fahrenheit. Solve for C in mended daily allowance of pro-
terms of F. tein. If one serving of Wheats
contains 8 grams of protein,
what is the recommended daily
20. SAT Scores. The average SAT allowance for protein?
verbal scores at a high school
have been increasing since new 22. Voting. In a presidential elec-
graduation requirements were tion, only 54% of the registered
inaugurated in 1980. y repre- If voters voted. If 42 million went
sents the average verbal score to the polls, how many regis-
and t the number of years since tered voters were there?
Answer to 1. x = $12,000 (at 8%) and 20,000 - x= $8000 (at 10%).

Practice Problem
1.5 Expressing Ideas with Graphs
Overview In this section, we introduce the rectangular (or Cartesian) coordinate system and learn to draw
the graphs of both linear equations and linear inequalities. It has been said that a picture is
worth a thousand words. Certainly managers, economists, and scientists often use graphs to
show ideas, rather than using equations. We must be proficient not only in interpreting graphs
but also in drawing graphs. This section is the foundation of graphical procedures emphasized
in this book.
We form a rectangular (or Cartesian) coordinate system with two per-

pendicular real-number lines, as shown in Figure 10. The two lines are called
coordinate axes. Traditionally, the horizontal line is called the x axis and the
vertical line is The point of intersection of the two lines is

called the y axis.
called the origin. The plane in which the two axes lie is called the coordinate
plane. The four parts into which the two axes divide the plane are called
quadrants. The four quadrants are labeled I, II, III, and IV, as shown in
Figure 10.
II
FIGURE 10
All six points seem to lie on the same it can be

straight line. In fact,
shown by ordered pairs
that this straight line contains all points represented
satisfying x + y = 4, and that each point on the line is represented by such a
pair, x + y = 4 is called a linear equation: that is, an equation of the form
ax + by = c, where a. b, and c are real numbers, a and b not both equal to
zero, is a linear equation.
FIGURE 11
EXAMPLE 29 Sketch the graph of p = q + 1, where p and q are real numbers.
SOLUTION For example, we choose to plot p along the vertical axis and q
this
along the horizontal axis. Now 2), (0, ), and (-1,0) are solutions of p
( 1 , 1
=
q + 1. If we represent these ordered pairs on a rectangular coordinate
system, we note (Figure 12) that a straight line contains all three points. We
might say that two of the points determine the line and the third point serves
as a check.
Pi
FIGURE 12
The y coordinate of the point where the line crosses the y axis is called
the y intercept. The x coordinate of the point where the line crosses the x
axis is called the x intercept. To find the y intercept, set x = in the equation
of a line and solve for y. To find the x intercept, set y = and solve for x. In
Figure 11 the x intercept is 4 and the y intercept is 4. In Figure 12 the q
intercept is -1 and the p intercept is 1.
EXAMPLE 30 Find the x intercept and the y intercept of the graph of 2 x - 3y = 6, and use
these to sketch the graph of the equation.
SOLUTION To find the x intercept, set y = 0. Then
2x - 3 = 6
x = 3 \x intercept]
To get the y intercept, set .v = 0. Then

2 - 3v = 6
v = -2 y intercept
The graph of 2x - 3y = 6 is shown in Figure 13.
FIGURE 13
If the graph is parallel to an axis, only one intercept is needed to

determine the graph of a linear equation, as seen in the following example.
EXAMPLE 31 Graph x = 4.
SOLUTION The x intercept is 4. Since x = 4 for each point on the graph, the
graph is parallel to the y axis, as seen in Figure 14.
FIGURE 14
Now let's look at one of many applications of the graphing of linear

equations. The intersection of the graph of a demand equation and the graph
of a supply equation is called the equilibrium point for supply and demand
(introduced in the preceding section). The point of intersection (.v, p) satis-
fies both the demand equation and the supply equation. Supply is said to
equal demand.
EXAMPLE 32 Suppose the demand equation is
320 - Ax
P =
Further, suppose the supply equation is p = 20.v. Find the point of equilib-
rium for these by graphing both the demand equation and the supply equa-
tion on the same coordinate system.
SOLUTION Note that when.v = 2, p 104; when x = 5, p = 100. Use these

two points to graph
320 - Ax
Use.v = 4,p = 80 and x = 6,p = 120 to graph p = 20.v as shown in Figure 15.
The intersection of the two graphs is at x = 5,p= 100. Thus, the equilibrium
price is $100. If the number of units is more than 5, the supply will exceed the
demand. If the number of units is less than 5, the demand will exceed the
supply.
Price
p= 20.v (supply)
(6. 120)
$120-
110"
100"
90-
Point of equilibrium
80--
70--
60--
50-
40--
30--
20--
10-
-\ 1
h
FIGURE 75
As seen in the preceding example, the graphing of equations is useful in

solving application problems. Just as important is the graphing of inequal-
ities.
As shown in Figure 16, a line divides the coordinate plane into two
parts called half-planes. Just as the points on a line are described by the
equation of the line, the points in a half-plane are described by means of an
inequality.
FIGURE 16
To illustrate, consider the line
v - 2.v - 1 =
shown in Figure 16. If (.v, y) is any point on the line, it will satisfy the
equation y - 2.x - 1 =0. If (*, y) is any point above the line, then it satisfies
y - 2x - 1 >
For example, (-1, 2) is above the line, and (-1, 2) satisfies the inequality
2 - 2(-l) - 1 = 3 >
In a similar manner, (0, 0) is in the solution set of
y - 2x - 1 <
since - - 1 < 0. This agrees with the fact that the half-plane below the
line consists of all points that satisfy
y - 2x - 1 <
Thus, an inequality using less than or greater than represents a half-plane,
and the corresponding line (< replaced by =) represents the boundary of the
half-plane. The following steps assist in finding the graph of an inequality.
Graphing Inequalities 1. Treat the inequality as an equation and plot points as if to graph the line.
2. If the relationship in the inequality is < or >, draw the line with dashes since the points
on the line are not solutions of the inequality but rather are on the boundary. If the
relationship is s or > draw a solid line since the points on the line are solutions.
3. By testing a point on each side of the line in the inequality, determine which side of the
line represents the set of solution points. (The origin is often a convenient point to
check if it is not on the boundary line.)
4. Shade the side of the half-plane that was found to contain the solution points.
EXAMPLE 33 Graph 3x + ly < 5.
solution 1 . Four points on the line represented by 3.v + 2_v = 5 are given in
Table 2.
TABLE 2
X
2. Sketch the line. Since no point on the line itself is a solution of 3x +

2y < 5, draw the line with dashes.
3. Determine which side of the line consists of points (x, y) such that 3x +
2y < 5. Select any point, say (3, 3), that is to the right of the line. Does the
inequality become a true statement if this point is substituted for the jc and
y values? Test it. Is 3(3) + 2(3) less than 5?
3(3) + 2(3) = 9 + 6=15

15 < 5
Therefore, points to the right of the line are not solutions of the inequal-
ity. Now test any point to the left of the line. Substitute the point ( 1 ,
- 1
to see if it gives a true statement. Is 3(1) + 2(-l) less than 5?
3(1) + 2(-l) = 3 + (-2) = 1
1 < 5
This point is one solution of the inequality.
4. Shade the half-plane to the left side of the broken line, forming the graph
in Figure 17>
S
\ 4.
\
\;
2-
H h
1 \2 3
\
\
\ 3x + ly = 5
FIGURE 17
EXAMPLE 34 Graph the solution set of y a x + I.
SOLUTION 1. The ordered pairs in Table 3 help to determine the line repre-
senting y = x + 1
TABLE 3
X
3. Find five solutions for each of (a) 2x - 3y = -6

the following equations and in-
(b)
(|)y
= 2.v + 4
equalities.
(c) x - 3v = 6
(a) 2.v+ 3v = 6
(d) 5jc - 3y = 15
(b) y <
x + 2
(c) 15 - 3y = 5x 7. Graph y = 2jc - 3 for - < x < 1
(d) 5x + 3y = 7 4. (Hint: The line extends from
x = -1 to x = 4.)
(e)
(|) x + y = 2y - 1
8. Graph y = 0.1 jc + 1.3 for

(f .v + v + 3 >
-10 < .v < 5.
)
4. Plot five points U, y) satisfying -2 <

= 3x + Draw 9. Graph y = -2x + 4 for
the expression y 1 .
x < 4.
a straight line through any two
of these points and observe the 10. Graph w < 2x - 4 for -1 <
results. x < 3.
5. Construct a graph for the points 11. Graph w < 3x 2 for -2

satisfying each equation or in- x< 2.
equality.
12. Give the equation of each of the
(a) y = 2x -3 axes in the coordinate plane.
(b) y > x + 2
13. Graph on the same coordinate
(c) 2x + 3y < 6
system y = mx + 2 for various
(d) 5* + 3y = 7
values of m. (Note the pattern
(e) 2x + y = 1
since we use it in the next sec-
(f) x + 1 =
tion.)
(g) y + 1 =
(h) y <
x - 1
14. Graph | + | < 5 for x > and
(i) y *>
(j) .v = 2x y > 0.
(k) 2s + t = 7
+ < 15. Draw a graph of the line that
(1) x y 9
passes through the points {(1,
6. Find the x and y intercepts and -1), (0, -3), (2, 1), (3, 3), (3/2,
graph each of the following 0), (- 1 , -5)}. Can you guess the
equations. equation of the line?
Applications
(Business and 16. Cost and Revenue. A revenue 17. Simple Interest. Draw a graph
Economics) function is given as R = 100.x. of the amount of money owed if
The cost function is defined by $200 is borrowed at 8% simple
C = 1000 + 80jc. Draw R and C interest for different periods.
on the same coordinate system Use the equation
and determine for what produc-
tion there is or is not a profit. A = $200(1 + 0.08/)
18. Depreciation. Let x in the equa- 20. Supply and Demand. Suppose
tion y = -60a- + 10,000 repre- the supply S (price per unit) for
sent months and y represent the the item in Exercise 19 is given
dollar value of a machine less by
depreciation.
5.v
=
(a) Prepare a table of values for
=
P T
-y 0, 5, 10, 15, 20, 25, 30,
(a) What is the price when .v =
35, 40.
3; when x = 9?
(b) Sketch the graph of the
(b) Graph this equation on the
equation, letting x assume
all real values greater than
same coordinate axes you
used in Exercise 19.
or equal to 0.
(c) Find the equilibrium point.
(d) When is the supply greater
than the demand?
19. Supply and Demand. Suppose
the demand D (price per unit) for 21. Supply and Demand. The de-
a certain item varies with the mand D and supply 5 for a cer-
number of units x so that tain commodity are given by
50 - 5.v 400 - 5.v 5_

P = ^T~ and p 2
(a) What is the price when (a) Graph the equation involv-
x = 0? ing Dand the equation in-
(b) What is the price when volving 5 on the same coor-
x = 4? dinate axes.
(c) What happens to the price (b) Find the equilibrium point.
when x = 10? (c) When is the supply less than
(d) Graph the equation. the demand?
and Life Sciences) 22. Height. Weight Chart. The desirable weight for a woman
Metro Health Club considers the who is 63, 66, or 69 inches tall.
following weights to be desirable
for women who are either 60 or
23. Population. The population of
72 inches
Montevallo seems to be increas-
tall.
ing linearly. In 1980, the popula-
Height x (in inches) 60 72 tion was 15,000. In 1986, it was

Weight y (in pounds) 100 146 20,000. Graph the linear equa-
tion that represents population
Draw a graph of a straight line in terms of time. Estimate the
through these two points and population in 1990 and 1994.
then estimate from the line the Hint: Let t = for 1980.
Answer to >
Practice Problem
1.6 Slopes and Linear Equations
Overview We have studied the solving and graphing of linear equations. Just as important is the ability to
develop an equation that accurately describes a situation. So in this section we develop equa-
tions of lines from data representing real-life situations. We define the slope of a line and use
this concept to find the graph of the line and also to find the equation of the line. In addition, we
learn to find the equation of a line through two points. This section is very important in your
accumulation of strategies to solve application problems.
Any two points in a plane can be considered as endpoints of some line

segment. Let (3, 2) and (6, 4) be endpoints of a line segment in Figure 19. If
we now construct through (3, 2) a line parallel to the x axis and through (6, 4)
a line parallel to the y axis, the lines meet at (6, 2). The change in x is 6 - 3 =
3 and the change in y is 4 2 = 2. The ratio of the change in y to the change
(6.4)
(3, 2)
H 1 1 1
h
FIGURE 19
The preceding concepts are important in discussing the inclination of a

line,which is measured by comparing the rise (the change in y: y2 - y\) to the
run (the change in x: x 2 - x\) as shown in Figure 20.
Slopes and Linear Equations 45
P 2 U 2 .v 2 )
Pi&i.y,) *Pz,(x2 ,y\)
FIGURE 20
Definition: Slope of a Line The ratio of the rise to the run of a line segment is called the slope of the line containing
the segment and is designated by the letter m. Thus, the slope of the line through P-,
(/,,/!) and P 2 (x2,y 2 )

is
/2~yi
x, *x 2
x2 -x.
Since the ratio of therise to the run is always constant, the slope of a
line is always the same no matter which two points are selected to compute
the slope.
If P 2 is to the right of P, , x2 - x t
will necessarily be positive, and the
slope will be positive or negative as y 2 - y\ is positive or negative. Thus,
positive slope indicates that a line rises to the right; negative slope indicates
that it falls to the right. That is, a slope of 2 means y increases by 2 when .v
increases by 1 . A slope of - 3 means y decreases by 3 when .v increases by 1
Since
y2 - yi -(yi - y2 _ ) \'i
- y2
X2 - -V| -Ui - X2 ) X[ - x2
P2 be to the right of P, is not necessary, and the order
the restriction that in
which the points are considered is immaterial in determining the slope.

The slope of the line segment from (1, 3) to (2, 3) is
3-3
2 - 1
This segment is parallel to the x axis. The slope of the line segment from
(2, 1) to (2, 4) is
or undefined
This segment is parallel to the y axis. In general, a line segment parallel to

the x axis has a slope of and a line segment parallel to the y axis has a slope
that is not defined. See Figure 21.
V
A linear equation may be solved for y, if d + 0, to obtain
y
T-v +
T
By letting in = -aid and b = -eld, the expression becomes
y = mx + b
Now consider the graph of y = mx (6 = example) when m has

for this
values i, 1,2, and 4 (see Figure 23). Since and (I A) are two points on
(0, 0)
the line v = (|).v, we note that the v value increases by \ as x increases by 1.

Thus, the slope is (I -0)/(l - 0) = .
Note also that m= in the equation. Since (0, 0) and (1,2) are points
on v = 2.v, the slope is (2 - 0)/( 1 - 0) = 2, and the coefficient of x in y = 2x is

likewise 2. Intuitively, you may have already decided that in y = mx + b, m
is the slope of the line.
FIGURE 23
If (.V| , vi) is a fixed point on a given straight line, and (.v, y) is any point
on the line, then the slope from (.v,, v,) to (.v, v) is
V| = m(x - .V|)
Since the coordinates x and y are variables denoting any point on the line,
the equation y- y\ = m(x - x\) represents the relationship between x and y.
Thus, the equation of the line with slope in passing through the fixed point
{x\ , y{) is y y\ = m(x - x\). A linear equation written in this form is said to
be in point-slope form. If the slope and one point of the line are known, then
the equation of the line can be easily obtained.
Point-Slope Form
EXAMPLE 36 Find the equation of the line through (2, 1) with a slope of 3.
SOLUTION
y - y\ = m(x - x t )
y - 1 = 3(x - 2)
y = 1 + 3* - 6
y = 3.v - 5
As a special case, the fixed point may be chosen to be the point where
the line crosses the y axis (the y intercept). The coordinates of this point are
usually written as (0, b). Then the equation of the line becomes y = + b, mx
the equation discussed earlier. The b in this equation is the value of y when
x = 0, or the y intercept.
Slope-Intercept Form If a line has a slope of m and a y intercept of b, then the equation of the line is given by
y = mx + b
EXAMPLE 37 If the slope of a line is 3 and the y intercept is 2, what is the equation of the
line?
SOLUTION Since m = 3 and b = 2, the equation is y = mx + bory = }>x + 2.
Sometimes, instead of being given a point and the slope, you are given
two points along the line. The two-point method of finding the equation of a
line consists of using these two points to determine the slope of the line and
then using the point-slope formula to establish the equation.
EXAMPLE 38 Find the equation of the line containing the points (2, 3) and ( 1, 4).
SOLUTION The two points can be used in either order to find the slope of the
line. One way is
4-3 I
The slope along with either of the fixed points can now be used to determine
the equation of the line; that is,
y - 3 = (~)(x - 2)
y-4 = (x+ 1)
Both of these may be simplified to give y = (-)* + ^ as the equation of the

line in slope-intercept form>
As a may be a horizontal
special case, the graph of a linear equation
line or a vertical (See Figure 14.) The equation of a horizontal line is the
line.
form y = b and has a slope of 0. The equation v = 1 is the equation of a
horizontal line as seen in Figure 24. By choosing two points on the line, say,
(2, 1) and (4, 1), the slope is seen to be 0. That is,
1-1
FIGURE 24
see that the slope is undefined. That is,
5-2 3
which does not exist.
y x = 3
5-- (3.5)
4--
3--
2-- -(3,2)
-4 -3 -2 -1
-1--
-2--
FIGURE25
Parallel and Perpendicular Lines

It can be proved that if two nonvertical lines are parallel, then they have the
same slope. Conversely, two lines with the same slope are parallel. In a like
manner, if two nonvertical lines are perpendicular, then the slope of one line
is the negative reciprocal of the other (m 2 = -1/mO. Conversely, if the slope
of one line is the negative reciprocal of the other, then the two lines are
perpendicular.
EXAMPLE 39 Find the line through (2, 1) perpendicular to x + 2y = 4.
SOLUTION The slope of .v + 2y = 4 is -=. \y = (~i)x + I
--"(-4)-* Slope of perpendicul ar line |
y - 1 = 2(x - 2) Point-slope form |
y = 2x - 3
The box that follows contains a summary of the types of equations

introduced in this section.
Slopes and Linear Equations
Summary 1. Point-slope form: Slope is m; line passes through (x,,y-i)
y-y, =m(x-x 1 )
2. Slope-intercept form: Slope is m; y intercept is b
y = mx + b
3. Horizontal line: y intercept is b; line has slope of
y = b
4. Vertical line: x intercept is h; slope of the line is undefined
x = b
Practice Problems > 1. For P,(l, 2) and P2 (4, 4), find y2 - y, , the change in v, and ,v 2 - jci , the
change in x.
2. Find the slope of the line through (-1, 2) and (3. -2).
3. Find the equation of the line that crosses the y axis at (0, -5) and has a
slope of 2.
4. Find the equation of the line through ( 1, 2) and (3, -2).
5. Find the equation of the horizontal line through (1, -2).
6. Find the y intercept of the line containing the points (2, 1) and (4, -2).
Answers to practice problems are found on page 54.
Exercise Set 1.6 Compute the slope or indicate that

the slope is undefined for the line
through each pair of points.
1.
In each of the following linear equa- 27. (1,3), (1, -2)

tions, what is What is the
the slope?
28. Find an equation of each line
y intercept of each? Graph each
with the following characteris-
equation.
tics:
12. y = 3x + 2
(a) The line contains the two
13. y + 2x - 1 = points (1, -3) and (4, 5).
(b) The line has a slope of -3

14. y = 3x - 1
and goes through the point
15. 2y = 1 -x (7, 1).
(c) The line has a slope of 1 and

4
16. v = goes through the point (-7,
1).
y- 1 (d) The line contains the two

17. x = points and
(0, 1) (4, 3).
(e) The has a y intercept of

line
18. 4x + 3y - 7 =
4 and a slope of 5.
19. 3x - 2>> = 5 (f ) The line has a y intercept of
6 and a slope of -3.
20. Classify the following state-
ments as either true or false. 29. Find an equation of the horizon-
tal line through (-4, -6).
(a) The slope of the y axis is 0.
(b) The line segment joining (a, 30. Find an equation of the vertical
b) and (c, b) is horizontal. line through (-5, 4).
(c) A line with a negative slope

31. Write the equation of the x axis.
rises to the right.
(d) A line that is almost vertical 32. Write the equation of the y axis.
has a slope close to 0.

33. Suppose the equation of a line is
written in the form
Find an equation of each line
through the given point with given
slope. a b
21. (1, 3), m = \ What is the x intercept? The y

intercept?
22. (0, 2), m = 1
34. Use the intercept form of the
-1
23. (-1, -2), m equation of a line (Exercise 33)
to find equations for lines with
24. (-3, 1), m = the following intercepts:
(a) x = 2, and y = -3
Find an equation of the line through
(b) x = 3, and y = 5
each of the following pairs of points.
35. Find the y intercept of the line
25. (1, 1), (2, 5)
that passes through the point (3,
26. (-1, 1), (2, 5) -2) with a slope of 2.
36. What is the slope of a line with a 39. In Exercise 23, find the equation
y intercept of -3 that passes of the line perpendicular to the
through the point (-4, 1)? given line through the point (2,
3).
37. In Exercise 21, find the equation
of the line parallel to the given 40. For Exercise 24, find the equa-
line through the point (2, 3). tion of the line through (2, 4)
perpendicular to the given line.
38. In Exercise 22, find the equation
of the line parallel to the given
line through (-1, -3).
Applications
(Business and 41. Depreciation. The decrease in per year, find its value at the
Economics) value of property over time is end of t years.

called depreciation of the prop- (b) What is the value of the ma-
erty. One accounting procedure chine at the end of 3 years?
for determining depreciation is (c) When is the value 0?
the straight-line method, or lin-
ear depreciation. In this proce- 42. Depreciation. A truck costing
dure the loss in value over a $12,000 has a useful life of 10

specified time is a given percent- years with a scrap value of $3000
age of the original value. For ex- at the end of this time. Find an
ample, if an item depreciates lin- equation representing value V

early at a rate of 5% per year, for linear depreciation in terms
then its loss in value at the end of time t
of 1 year is 0.05C, where C is the

43. Sales. Assume the amount of
cost. At the end of 4 years its
sales over time for the Cenco
loss in value is
Corporation is given by a linear
4(0.05)C = 0.20C equation. Suppose the amount
of sales was $150,000 in 1980
Let r be the annual rate of
the length
and $110,000 in 1984.
linear depreciation, /
of time in years, and C the origi- (a) Let t = in 1980; find an ex-
nal cost of an item. At the end of pression for the amount of
1 year, the item depreciates in sales S in terms of time /.
value by Cr. At the end of t (b) What was the amount of

years, the item has depreciated sales in 1983?
by Crt. Therefore, its current (c) Estimate the amount of sales
value (original cost minus depre- in 1986.
ciation) is
44. Marginal Cost, Revenue, and

V = C-Crt or V=C(l-rt)
Profit. We will learn later that
(a) If a machine costing $10,000 in a linear equation representing

depreciates at a rate of 5% cost, such as C = mx + b, the
slope m is called marginal cost. 45. Profit Function. Given the profit
In revenue equations, such as function P = 25* + 72,
R = mx + c, the slope m is
(a) What is the slope?
called the marginal revenue. In
(b) Interpret the slope.
profit equations, such as P =
(c) What is the P intercept?
mx + d, the slope m is called the
= (d) What is the marginal profit?
marginal profit. that C Given
(e) Interpret the marginal profit.
400 + 3x and R = 7x+ 100, find
the marginal cost, the marginal
revenue, and the marginal profit.
Hint: P = R - C.
and Life Sciences) 46. Pollution. In a certain indus- gin and t as the time in hours
trial city, it is believed that the after 7:00 a.m. to find the linear
pollution count increases lin- equation representing the pollu-
early from 7:00 a.m. to 2:00 p.m. tion count in terms of time. Pre-
At 8:00 a.m., the pollution count dict the pollutioncount at 11:00
is 140. At 10:00 a.m., the count a.m. and at 1:00 p.m.
is 200. Use 7:00 a.m. as the ori-
Answers to Practice > 1. y2 - y t

= 2; x2 - x = {
3
Problems __ ,
2. The slope is 1. ,
3. The equation is y = 2x - 5.
4. y = -x + 1
5. y = -2
6. The y intercept is 4.
Extended Application
Budget Lines 1
At the ends of many of the chapters of this book is presented material from
advanced textbooks, applying the concepts you are studying. Here we con-
sider an application of lines, slopes, and inequalities as found in a microeco-
nomics textbook. The only purpose of this example is to show you that what
you have learned can be used in later courses.
In real life, there are many goods from which a consumer can choose.
In this study a consumer uses only two goods. (A study of only two goods is
1
Varian, Hal R. Intermediate Microeconomics: A Modern Approach, 1987, W. W. Norton and
Company. New York, pp. 21-25.
Summary and Review 55
much more general than it seems, and provides much valuable information.)
The number pair (.v, y) is called a consumption bundle, where .v represents
the amount of good 1, and y the amount of good 2, the consumer chooses.
Let/?, be the price per unit of good and p 2 the price per unit of good 2. p\x
1
is the amount of money spent on good 1, and p 2 y, the amount of money
spent on good 2. Let M be the amount of money the consumer has to spend.
Then the budget constraint is
P\.x + p2 y M
The following graphs of the line p,.v + p2 y = M introduce the properties of a
budget set and provide an interesting application of slopes and lines, as well
as the solution sets for inequalities.
V
Graphing (34) Origin (34)

Half-plane (38) Point of equilibrium (37)
Inequalities (23) Polya's four steps (2)
Integer (11) Proper subset (8)
Intercepts (36) Rational numbers (11)
Inverse of addition (14) Real numbers (11)
Inverse of multiplication (14) Rise (44)
Irrational numbers (11) Run (44)
Less than (23) Set (8)
Greater than (23) Slope (45)
Linear model (28) Suggestions for understanding a
Multiplicative identity (12) problem (29)
Multiplicative inverse (12) Supply and demand (28)
Natural numbers (8) Whole numbers (11)
Ordered pair (34)
Important -(-x) = X Slope of line segment from (jci .
Formulas = > y,) to (x 2 , y2 ) is

\x\ x if x
|.v| = x if v <
Xl ~ X\
x + (-y) = x - y if
|x| > M Properties of Equalities

= -(y - x) if
If a = b, then ac = bc,ca =
|x| < \y\ cb, and ale = blc if c +
x + (-x) =
-x + (-y) = -U + y)
Properties of Inequalities
x(y) = -xy (unlike

If < b, then a c < b c;
a
ca < cb and ale < blc when
signs)
r > 0;
-x(-y) = xy (like signs) ca > cb and ale > blc when
c < 0.
= P c * 0, q *
cq q
Equations of Lines
P + r
Point slope: y - yi = m{x - X\)
q q q
Slope intercept: y = mx + b
p +
q
^
s
ps + qr
qs
Horizontal: y = b
Vertical: x = h
Summary and Review 57
Review Exercises Write the following in mathe-

matical language.
(a) Three times a number less 6

equals 21.
(b) Eight less than 4 times a
number equals 61.
(c) A number less 42 is always
less than 20.
Let A = {-2, - \, -V2, 0,
?=, V2, 2}. List the elements
of A in the following sets.
(a) Set of natural numbers

(b) Set of integers
(c) Set of rational numbers
(d) Set of irrational numbers
(e) Set of negative numbers
(f Set of real numbers
Solve the following equations

and check your answer.
(a) 6.v
(a) 2x - 6 = 4 21. A will provides that an estate is
to be divided among a wife and

(b) 4 = -5
3 three children. The wife is to re-
x + 2 ceive $4000, and the remainder
(c) = -1
is be divided 4 ways, with the
to
(d) 5.v + 4 = 14 wife receiving twice that re-

ceived by each child. If the es-
13. Solve the following inequalities,
tate is valued at $24,000, how
giving the reason for each step.
much does each receive?
(a) 2x - 6 < 4
Pam holding
22. is $3.25 in her
(b) 4 > 5 hand. If she has twice as many
nickels as quarters and 1 more
(c) -3jc + 7 < -2 dime than quarters, how many
(d) ^+ 5 > 7 coins does she hold? How many
coins of each type does she
14. Using the illustration in the Ex- have?

tended Application, write the 23. To produce a new record, Shoe-
equation of the linepi.v + p 2 y = string Recording Enterprises
M in the slope-intercept form of must make an initial investment
the equation of a line. of $600. Then each record pro-
duced costs an additional $2 for
15. In Exercise 14, what is the
manufacturing, sales, and royal-
slope, the x intercept, and the y
intercept of the line?
ties. How many copies of a rec-
ord can manufactured if
be
16. In Exercise 14, what does p x t Shoestring Recording Enter-
represent? p 2 y^ prises has available $1500 of
capital?
17. In Exercise 14, explain how we
can get 24. Canoeing enthusiasts find that
they can paddle upstream at 4
M M I
kilometers per hour and down-
stream at 12 kilometers per
from the original equation.
hour. How long can they paddle
18. Solve and graph the following upstream if they must return to
inequality: |3 jc + 1| < 5. the same spot 4 hours after de-
parting?
19. Solve for t in at + b = 2t.
20. Solve for x using your calcula-

25. Two backpackers start toward
each other from 2 points on the
tor.
Appalachian Trail. Initially they
(a) 1.233* - 4.007 = 32.1 - are 12 kilometers apart. If one
0.0043.x walks 4 kilometers per hour, and
the other walks 2 kilometers per
(b) j^rz - 14.745 = 0.001 x +
hour, how long will it be until
2.1141 they meet?
Chapter Test 59
26. If the $180 you spend each 23,000 automobile-accident fa-

month on room and board repre- talities occurred during daylight
sents 20% of your montly in- hours, how many such fatalities
come, what is your monthly in- occurred that year?
come?
29. Lynn took her 100-point history
27. An item sells for $1.45. If the final yesterday. She missed
store uses a 40% mark-up to de- twice as many questions on the
termine selling price, how much multiple-choice section as on the
did the item cost? What is the true-false section. She missed a
selling price of an item that costs total of 9 questions. Each true-
$2.00? false question was worth 1
point; each multiple-choice

28. According to the National
question was worth 5 points.
Safety Council, about 60% of all
She earned 2 points on the bo-
automobile-accident fatalities
nus question. What was her
for a year occurred during day-
grade on the test?
light hours. If, during the year,
Chapter Test
1. Draw the graph of 2x + 3y = 6.
2. Express Alice's age in mathematical terms, letting n be Tom's age now.

Aliceis 3 years younger than 5 times Tom's age year from now. 1
Simplify g +
-f){
4. Find the slope of the line that contains (2, 5) and (-1, 6).
5. Draw the graph of .v + 3v > 9.
6. Solve -2x + 4 < 6.
7. Find the equation of the line through (2, 3) perpendicular to x = 7.
8. Find the equation of the line through (2, 3) parallel to 2.v + 5v = 7.
\2x + 41
9. Solve = 5.
10. Solve 5 for r.
11. If you spend 20% of your monthly income on food, what is your monthly
income if you spend $300 on food?
12. A woman has an annual income of $13,000 from two investments. She
has $30,000 more invested at 10% than she has invested 12%. How much
does she have invested at each rate?
13. If U = {3, 4, 5, 6, 7, 8, 9, 10} and A = {4, 5, 7, 8}, find the complement

of A.
14. Find the y intercept of the line containing (1,2) and ( 1, 6).
Linear Algebra
In this chapter you will be considering one of the most useful tools of man-
agers and scientists. With the increased use of computers and
large quanti-
ties of data, a convenient scheme was necessary

for storing information in
the computer. One such tool was to arrange the data in the
form of a matrix.
Business executives, government agencies, social scientists, and life
scien-
tists all use matrices.

we observed that the solution set of a linear equation such
In Chapter 1
as
ax +
consists of an infinite number of ordered pairs of real numbers. In fact, we

graphed such linear equations in two unknowns and observed that the graphs
were straight lines. often useful to determine if equations have common

It is
solutions. The equations are then referred to as a system of equations, and

the ordered pairs that satisfy all equations are the solutions
to the system. We
begin our study of solving systems of equations by looking for graphical
solutions; then we review algebraic procedures. To solve
systems involving
a large number of variables, we use matrices in two ways.
We conclude the
chapter with an interesting example called input-output analysis.
61
62 Chapter 2 Linear Algebra
2.1 Getting Acquainted with Matrices
Overview In this section you will learn that a matrix is a very simple, yet enormously valuable, concept. We
introduce the concept with an inventory matrix. (Just think how many companies and stores
keep inventories!) You will learn to do basic operations on matrices. (This is sometimes called
matrix algebra.) Also, we illustrate in this section why matrices provide a convenient and con-
cise way to store data and thus why they are used so frequently in computer applications. In this
section you will study
The meaning of a matrix

The size of a matrix

Addition of matrices

Scalar multiplication

The negative of a matrix
A matrix is a rectangular array of objects (most often numbers).
EXAMPLE 1 The following rectangular arrays are matrices:
2 1
2 1 4 2 3
3 2 [10 3 1]
3 2 7 1 5
4 5
An inventory matrix is a matrix used to store inventory data.
EXAMPLE 2 Suppose Pizitz Department Store has 3 warehouses to store equipment for
sports. Suppose there are 10 tennis rackets, 17 baseball bats, 11 footballs,
and 5 pairs of handball gloves in warehouse A. In warehouse B, there are 12
tennis rackets, 11 baseball bats, 15 footballs, and 7 pairs of handball gloves.
Also, in warehouse C there are 20 tennis rackets, 12 baseball bats, 32 foot-
balls, and 20 pairs of handball gloves. This information could be expressed in
table form as
TABLE 1
Getting Acquainted with Matrices 63
Matrices are classified by their size (sometimes called dimension or

order), thatis, by the number of rows and columns they have, with the
number of rows being given first. In Example 1, the first matrix has two rows
and three columns. It is a 2 x 3 matrix (read "2 by 3"). The second matrix
has three rows and two columns, or is 3 x 2. The third matrix is 2 x 2; the
fourth matrix is 1 x 3; and Example 2 the matrix is 4 x 3.

in
A matrix with the same number of rows and columns is called a square
matrix. The following are square matrices.
3 7 2
2
5 -3
1
and 4-16
2 1
3x3
A matrix may in general have m rows and n columns and be written as
a\\ an i3 a\ n
2I #22 #23
\_a m \ a,i am T, ... a,
The ism X n, where m represents the number of rows and

size of this matrix
n the number of columns. The objects in the matrix are called elements of the
matrix. Usually, the elements are numbers. Notice that the first part of the
subscript notation for each element gives the which the element lies, row in
and the second part gives the column location. For example, a 35 is the
element in the third row and fifth column. The general m x. n matrix is
sometimes written as [a,-,-], where a y represents each element.
EXAMPLE 3 2 3 4
A =
1 5 7
is a 2 x 3 matrix. Its elements are a,, = 2, a ]2 = 3, i 3 = 4, a2 i

= 1. #22 = 5,
and fl23 = 7.
A matrix consisting of a single row of elements is called a row matrix.

Similarly, a matrix consisting of a single column is called a column matrix.
EXAMPLE 4 A = [1 3 7] is called a row matrix with elements a,, = 1, a l2 = 3, and

I3 = 7.
7"
-1
is called a column matrix with b\ 7,6m 1, and 6 3 = , 4.
When new mathematical objects are introduced, one of the first ques-
tions to be answered is, "When are two of these objects equal?" In our case
the question would be "When are two matrices equal?" The answer is fairly
easy:
Definition: Equal Matrices Two matrices are equal if and only if they have the same size (same number of rows and
columns) and if all corresponding elements are equal.
EXAMPLE 5
L5 2} U? 2]
These two matrices are equal because they have the same size (that is, they
are both 2 x 3), and all corresponding elements are equal:
14
2 =
EXAMPLE 6 3 2 1 3 2 10
5 2 5 2
These two matrices are not equal because they are not the same size; that is,
the first matrix is 2 x 3, and the second is 2 x 4.
If A T is matrix A with the rows and columns interchanged, then A r is

called the transpose of matrix A. For example,
'2
f
3 2
1 4
is the transpose of
2 3 1
1 2 4
The first column of AT is the first row of A. The second row of A is the
second column of A7 .
Now let's begin our study of matrix algebra with addition. Two matri-
ces can be added if they have the same size. We find the sum of two matrices
by adding corresponding elements.
Definition: Sum of The sum of two matrices A and B of the same size is the matrix with elements that are the
Two Matrices sum of the corresponding elements of A and B; that is, the entry in the /th row and /th
column is a,, + bn .
For instance, the sum of
1 -2 1
and
2 3 1 2
1 -2 1 1 + (-2) + 1 -1 1
2 3 1 2 2 + 1 3 + 2 3 5
Note that according to the preceding definition, matrices can be added

only when they have the same size. For example,
3
A matrix with each element equal to is a zero matrix, also denoted by

0. For example,
"0
is a 2 x 3 zero matrix. Note that
2 3 1] [0 2 3 1
-10 5 -10 5
We will be interested in two kinds of mathematical products involving matri-

ces: (1) the product of a real number and a matrix and (2) the product of two
matrices. For the product of a real number and a matrix, the real number is
multiplied by each element of the matrix. For example,
3 1 2-3 2-1 6 2
2 2-0 2-2 4
In working with matrices, a real number is called a scalar.
Definition: Scalar Product The product of a real number c and a matrix A = [a,
7]
is the matrix [ca,7 ]. That is,
c[a,7 ] = [ca,
7]
EXAMPLE 9 1 -1 2 5-1 5
(-1) 5
2 -5 10
3 4 5-3 5-0 5-4 20
The following properties can be easily developed:
Properties of Scalar If c and d are any real numbers:

Multiplication 1. c(dA) = (cd)A
2. cA + dA = (c + d)A
3. c(A + B) = cA + cB
The preceding definition of the product of a constant and a matrix can

be used to find the negative of a matrix. One matrix is the negative of another
if their sum is a zero matrix.
Definition: Negative The negative of a matrix A = [a, denoted by -A, is the product of -1 and A. That is,
7 ],
of a Matrix -A = r-aiil.
EXAMPLE 10 If
2 -1
A = 4 -2
-3 3
then
-2 1
-4 2
3 -3
3. Perform the following computation.
2 31 ro 1
5 lJ L2 1
4. Find the negative of the matrix
r-i i 2i
L-2 4 -5J
Exercise Set 2.1 Determine the size of each matrix

and find the element requested.
1. A = U4 1 31
L2 -1 OJ
2. B =
3. C =
Find the transpose of each of the fol-

lowing matrices:
18.
3
19.
20.
28. Markov Matrices. If knowledge The second row represents what

of a current state of a situation happened to the customers of
aids in the prediction of the next store B. Interpret the second
state, the next state of the situa- row.
tionis often modeled by a Mar-
29. Marketing. Suppose two news-
kov matrix. Two stores in a city,
papers, the Star and the Times,
A and B, are competing for cus- are competing for customers.
tomers. The elements in the first
During the last year it was found
row of the matrix represent the
that the Star kept60% of its cus-
proportion of customers who re-
tomers and 40% to the
lost
main with A
and who change to
Times. The Times kept 70% of
B. A survey shows that last year
its customers and lost 30% to the
store A lost 20% of its customers
Star. Write a matrix for this sur-
to B and kept 80% (first row of
vey. (See Exercise 28.)
the matrix).
Next State
A B
Present A [ -80 .20

State b L -24 .76
and Life Sciences) 30. Demographics. Can you write a the following matrix.
matrix showing the changes in Offspring
three classifications of employ- R H D
ment? The first row would rep-
.75 .25
resent changes from business to
.25 .50 .25
the three classifications and
.75 .25
would be [.75 .15 .10]. That
is, 75% of those in business re- Assume that an inherited trait is
main in business. Now complete governed by a pair of genes of

the matrix.
From business (or industry) to business (or industry) 75%

From business (or industry) to unemployment 15%
From business (or industry) to self-employment 10%
From unemployment to business (or industry) 25%
From unemployment to unemployment 60%
From unemployment to self-employment 15%
From self-employment to business (or industry) 5%
From self-employment to unemployment 5%
From self-employment to self-employment 90%
31. Genetics. The characteristics type G or g. Possible combina-

of an inherited trait are given by tions are gg (recessive and de-
noted by/?), gG (hybrid, //), and 33. Communication. Matrices are

GG (dominant, D). In the first used in communication-network
row of the matrix, 75% of the studies and to show sociological
offspring of recessive parents relationships. A reception is in-
are recessive, 25% are hybrid, dicated by 1 and no reception

and none is dominant. Interpret by 0.
the remainder of the matrix.
32. Transportation. The following

figure shows the routes of direct
air flights among five cities.
1 1 Matrix Multiplication
Overview In this section we consider the product of two matrices. Do not confuse matrix multiplication
(multiplying two matrices) with the multiplication of a matrix by a constant as considered in the
preceding section. You will discover that matrix multiplication is a bit more complicated than
addition of matrices and that matrix multiplication satisfies few of the properties we associate
with multiplication. In this section we will study
The dot product of two matrices

The definition of matrix multiplication
The arithmetic of matrix multiplication
Applications of matrix
multiplication
A gallon of vanilla ice cream uses 3 quarts of milk, 1 pint of cream, 2

cups of sugar, and 6 fluid ounces of vanilla extract. We write this information
as a row matrix.
[3 12 6]
Suppose milk costs $0.65 per quart, cream $0.50 per pint, sugar $0.12 per
cup, and vanilla extract $0.14 per fluid ounce. We write the costs as a
column matrix.
"0.65
0.50
0.12
L0.14J
We desire to multiply the row matrix times the column matrix to get the total
cost of a gallon of ice cream. Your "common-sense" multiplication actually
defines the dot product of two matrices.
3(0.65) + 1(0.50) + 2(0.12) + 6(0.14) = $3.53
We begin our study of matrix multiplication by considering the dot

product of the row matrix [1-2 3] and the column matrix
2
-1
The dot product of these two matrices is
[1 -2 3] = 1(2) + (-2K-1) + 3(1) = 7
The answer is a real number that is the sum of the products of corresponding
entries (first element in a row times first element in a column plus second
element in a row times second element in a column, etc.).
Matrix Multiplication 73
Definition: Dot Product The dot product of a 1 x n row matrix and an n x 1 column matrix is defined by
a, a2 . . . a] = aibi + 8262 + + a n b A real number
The dot between the two matrices indicates the dot product and distin-
guishes this product from matrix multiplication, to be defined later in this

section.
EXAMPLE 13 2
[1 2 3] -1 = 1(2) + 2(-l) + 3(1) = 3
1
Size Size
1 x 3 3 x 1
To obtain the dot product,

these numbers as paired
must be the same.
EXAMPLE 14
[3 2] = 3(-l) + 2(1) = -1
Size Size
1x2 2x1
I , 1
Same
EXAMPLE 15
[2 3 -1 4]
Size Size
1 x 4 3 x 1
L J
Not the sa me (dot product not defined )
Now let's use the ideas of the preceding examples to define the product
of two matrices.
Definition: Product of The product of the m x p matrix A and p x n matrix B is the m x n matrix AB whose
Matrices element in the /th row and /th column is the dot product of the fth row of A and the /'th
column of B.
Consider the multiplication of two matrices such as
and
To obtain our product matrix, we find the element c\ \ in the first row and first
column of the product matrix by finding the dot product of the first row
[3 1 2] of the first matrix (matrix on the left) and the first column
1
4
-2
of the second matrix (matrix on the right). In this example,
[3 1 2] 4 (3)(1) + (1)(4) + (2)(-2) = 3 + 4-4 = 3

-2
This element goes in the first row, first column of the product matrix. If the
product matrix denoted by Cy, then c\\ = 3. Next, to find en, we take the
is
dot product of the first row of the left-hand matrix with the second column of
the right-hand matrix.
[3 1 2] 3(0) + 1(6) + 2(3) = + 6 + 6= 12
This element goes in the first row, second column of the product matrix:
en = 12. Then we take the dot product of the second row of the matrix on the
left side and the first column of the matrix on the right.
t-1 5] (-l)d) + 0(4) + 5(-2) = -1 + - 10 = -11
This element goes in the second row, first column of the product matrix:
c 2 = -11. Then we take the dot product of the second row of the left-hand
\
matrix with the second column of the right-hand matrix.
[-1 5] (-1K0) + 0(6) + 5(3) = + + 15 = 15

This element goes in the second row, second column of the product matrix:
c2 2 = 15. Thus,
1
3
4
-11
-2
Notice that the product of two matrices is not defined unless the num-
ber of columns of the first matrix is the same as the number of rows of the
second matrix. Another important item to note in the definition is the size of
the matrices involved. A 3 x 2 matrix multiplied by a 2 x 3 matrix gives a
3x3 matrix, whereas, if the order is reversed, the product of a 2 x 3 matrix
with a 3 x 2 matrix gives a 2 x 2 matrix. Thus, the order of multiplication is
important when multiplying matrices.
Matrix A Matrix B
2 x 3 3x2
Must match
Size of product AB
The following example indicates that the commutative property of mul-

tiplication does not hold for all pairs of matrices. Many times, the product
will not be defined if the order of multiplication is reversed. For example, a
2x3 matrix times a 3 x 5 matrix results in a 2 x 5 matrix. However,
multiplication is not defined for a 3 x 5 matrix times a 2 x 3 matrix. Even
when both products are defined, AB is often not the same as BA.
In the example given earlier,
3 1 2 3 12 a 2i = -1
1 + 0-4 + 5(-2)
4 6
-10 5
-2 3
11 15 = -II
If the order is reversed, the product becomes
3 1 2
1 2 a2 ,
= 4 3 + 6(-l)
4 38
-1 5 = 6
-9 -2 11
Hence, the commutative property does not hold for the product of two
matrices.
In much of the work with matrices in this book, we will use only square
matrices; that is, matrices where the number of rows and the number of
columns are the same. The product of square matrices of the same size
always exists. In particular, we denote the products
(A) (A) as A 2
(A) (A) (A) as A 3
(A) (A) . . . (A) as A"
n factors
For example, if
2 3
A=
L-4 1
then
2 3] r 2 3
A = -4
2
1 -4 12
1
The identity matrix for multiplication is very important in the sections

that follow. Just as 1 is the identity for multiplication of real numbers (that is,
1
6 = 6
1 = 6 or 1
x = x
1 = x), we define
1
I =
1
as the identity matrix for 2 x 2 matrices. For example, I A = AI = A, or if
Likewise,
Note that the identity matrix is a square matrix.

EXAMPLE 16 At the Barber Ice Cream Company a matrix lists the various quantities of
milk (in quarts), cream (in pints), sugar (in cups), vanilla extract (in fluid
ounces), and baking chocolate (in ounces) to make a gallon of different
flavors of ice cream (only two of which are listed).
Cream
Vanilla
Chocolate
Milk
3
3
12
1
Sugar
2i
Vanilla
6
1
Chocolate
If milk costs $0.65 per quart, cream $0.50 per pint, sugar $0.12 per cup,
vanilla extract $0.14 per fluid ounce, and chocolate $0.20 per ounce, find the
cost of a gallon of vanilla and the cost of a gallon of chocolate ice cream.
SOLUTION
0.65
0.50
3 12 6
0.12
3.53
3 1 24 1 5 3.89
0.14
0.20
Thus, vanilla ice cream costs $3.53 per gallon to make (not including the cost
of labor) and chocolate ice cream costs $3.89 per gallon.
Matrices can also be used to replace a system of equations by a single

matrix equation.
EXAMPLE 17 Write the following system of equations in matrix form.
3*, + 4*> - 2;t 3 + .v4 = 5

xi + 3;c 2 - 2*4 = 7
2*i + 3.v 2 + xt,
- *4 =
*i - *2 ~~
*3 + 3,v4 = -2
SOLUTION This set of four equations in four unknowns can be replaced by the
single matrix equation
3 4
1
2
L 1
as can be seen by multiplying the two matrices on the left to obtain

Since the two matrices are equal, the definition of equality demands
that their corresponding elements be equal. Hence, we obtain the given
system of equations. If
A =
Suppose A is a 2 x 3 matrix, B is a 3 "1 5

x 4 matrix, C is a 4 x 4 matrix, and 22.
D is a 4 x 3 matrix. Determine
whether or not each of the following
products is defined, and if it is de-
23.
fined, specify the size of the matrix
answer:
7. AB
24.
8. BA
9. AC
10. BC
11. CB 25.
12. CD
13. B(CD)
14. A(BC)
26.
15. (AB)D
Compute the matrix answer if de-

27.
fined.
2 1 6
16. [1 3]
1 -1 28.
3 1
17. -4 1
-1 1
2
18. [1 3 -1] 1
3 1
A is the matrix of coefficients, and B Follow the instructions for Exercises

is the column matrix of constant 33-38 where
terms:
33. Ax + 3y = 7 X =
6x - y = 10
39. 6x + 3y + lz = 2
34. 3x - y = 7
5x + Ay + z = 5
5x + 2y = -1
2jc - y + z = 6
35. y = Ax + 3
40. y + z = 6
x = 2y - 1
x = 7 - y
36. y + x = 2
x = z - 6
y = A
41. jc = 4
37. x = A
y = 7
x + 2y = 1
z = 9
38. x = 6
y = A
Applications
(Business and
Economics) 42. Stocks. In the previous exer-
Consumer
cise set we had the following ta-
Bonds Mortgages Loans
ble of the purchases of stock by
an investment club. Using matri-
City A
City B
ces, find the total cost of all
City C
stocks.
end of the period. For example, Check this multiplication and

then see if you can give an inter-
.80
[.60 .40] pretation of the meaning of
.24
[.576 .424].
[.576 .424]
and Life Sciences) 45. Demographics. Find A 2
for Ex- Creek by dumping industrial
ercise 30, Set 2.1, and interpret wastes from 4 manufacturing
the elements of the product as processes. Three pollutants are
the percentages after 2 periods found in the creek. The follow-
of time. ing quantities of pollutants, ex-
pressed in milliliters, are found
46. Genetics. Find A 2
for Exercise
per 1000 liters of water for each
31 Set 2.1, and interpret the ele-
,
process.
ments of the product as traits
from parents to grandchildren.
47. Transportation. The square of

the matrix in Exercise 32, Set
2. 1 , gives one-stop flights. A2 in
the product indicates 2 one-stop
flights.What cities are con-
nected with 2 one-stop flights?
48. Communications. Interpret A 2
in Exercise 33, Set 2.1, as a

transmission from one station to
another through an intermediate
station. Describe the elements
of A 2
.
49. Pollution. Lownes Chemical is
accused of polluting Bacon
-5
Answers
Practice Problems
to 1.
r 10 2
'
U 1} I
1
Ml 2 -1 2 1 3 1
1 3 1
we solve systems of equations using matrices. Of particular interest is a suggested procedure
for calculator solutions.
First let's consider a system of two equations in two unknowns such as
2x-3y= 1
x + y = 3
The solutions of any equation in two unknowns is the set of ordered

pairs that satisfy the equation. A solution of a system of two equations is
defined to be a common solution of the individual equations. Such solutions
are called simultaneous solutions since each ordered pair satisfies both equa-
tions simultaneously. A solution of the system above is the ordered pair
(2, 1). Check to see that this ordered pair satisfies both equations.
Our first procedure for obtaining such a solution is to solve the system
graphically. The graph of the solutions of each linear equation in two un-
knowns is a straight line. Suppose now that two such equations are graphed
on the same coordinate system. The solution of the system of two
linear equations in two unknowns is then given by the intersection of the two
lines.
EXAMPLE 18 Solve the following system graphically.
3.v + y = 3
x + ly = -4
SOLUTION The first equation is satisfied by infinitely many ordered pairs,

three of which are (0, 3), (1, 0), and some ordered pairs
(3, -6). Likewise,
that satisfy x + 2y = -4 are (-4, 0), (0, -2), and (2, -3). The graphs of
these two equations are given in Figure 1. The intersection of the two lines in
Figure 1 seems to be the point (2, -3). We can see whether or not this
ordered pair a solution of the system of equations by checking it in each
is
equation. Substituting x = 2 and y = -3 in the first equation gives
Check: 3(2) + (-3) = 3
6-3 = 3
Substituting x = 2 and y = -3 in the second equation gives
Check: (2) + 2(-3) = -4
2 - 6 = -4
Hence, (2, -3) is a solution of the system.
Geometrically, we are confronted with three possibilities for the

straight-line graphs of equations in a system of two linear equations in two
unknowns.
Solving Systems of Equations 83
FIGURE 1
Intersection of Lines One of these possibilities must occur for the graph of two lines in a plane:
1. The two lines intersect at exactly one point.
2. The two lines coincide.
3. The two lines are parallel.
Correspondingly, these possibilities lead to the conclusion that one and

only one of the following is true for two linear equations in two unknowns x
and y:
1. The two lists of solutions contain exactly one common ordered pair,
called the solution of the system.
2. The two lists of solutions are identical.
3. The two lists of solutions contain no common ordered pairs.
Each of these possibilities is illustrated in the following example.
EXAMPLE 19 Find the solution set of each of the following systems of equations,
(a) 6.v + 2y = 8 (b) 3.v + y = 3 (c) 3* + y = 3
3.v - ly =1 6.x + 2y = 6 6x + 2y = 12
In (a), the solution is ( 1 , 1 ). There is an infinite number of solutions (all
the points of the line) in (b), since all the solutions of one equation are
SOLUTION
FIGURE 2 (c)
solutions of the other. One equation can be obtained from the other by
multiplying both sides of the equation by a constant.
2(hx + y = 3) gives 6x + 2y = 6
There are no common ordered pairs in (c) since the lines are parallel.
Although the graphic solution of a system of two linear equations with

two variables gives an excellent picture of the relationship between the two
variables, the method is time-consuming and may not be accurate if the
numbers that compose the ordered pairs in the solution set are not integers.
Furthermore, they do not easily generalize to larger systems of equations.
Consequently, algebraic methods for solving the system are often more prac-
tical. We will study how to algebraically obtain a second system of equations
that is equivalent (that is, it has the same solution) to the given system.
One such method, method, is used ex-
called the addition-subtraction
tensively. Thismethod seeks to combine the equations in such a way that
one of the unknowns appears with a zero coefficient. This can happen by
making the coefficients of the selected variable additive inverses in the two
equations; then by addition of the two equations, the coefficient of this
variable becomes 0. We use this procedure now to solve the following sys-
tem of equations:
4x + 2y = 8 (1) Let's eliminate x. What multiplied by

3.v and what multiplied by 4.x will
3-v ~~ =
y make the coefficients inverses?
In this illustration we multiply both sides of the second equation by -4 and

both sides of the first equation by 3.
12.x + 6y = 24 (1) This system is equivalent to the

preceding system; that is, it has
12.x + 4y = tne same solution.
If we add equations 1 and 2 term by term, we obtain
lOy = 20 or y = 2
Substituting y = 2 in equation 1 yields
4.x + 2(2) = 8
4.x = 4
x = 1
Check: 4(1) + 2(2) = 8 (1)
8 = 8
3(1) -2=1 (2)
1 = 1
So the solution is (1, 2).

EXAMPLE 20 Find the solution of the system
3.v + y = 3 (1)
x + 2y = -4 (2)
SOLUTION In order to make the coefficients of y additive inverses in the two

equations, multiply each term of the first equation by -2 to obtain
-6x - 2y = -6 (1)
Add to this equation the like terms of the second equation:
-6x -2y= -6 (1)
x + 2y = -4 (2)
-5x = -10
x=2
Substituting x = 2 into the first equation gives
3(2) + y = 3 or y = -3
The solution is (2, -3), and we can check that this point lies on both lines by
substituting its coordinates into both equations. Note that this is the same
solution as was obtained by graphical procedures.
EXAMPLE 21 Find all the solutions of the system
3x + y = 3 (1)
6x + 2v = 6 (2)
SOLUTION Multiply both sides of the first equation by -2 to obtain
-6x -2v = -6 (1)
Add this equation to the second equation to obtain
If (jc, y) satisfies
3* + y = 3 (1)
then multiplying by 2 gives
2(3.v + y) = 2 3 (1)
or
6x + 2y = 6 (2)
Hence, any point that satisfies the first equation will satisfy the second
equation. That is, the graphs of the two equations coincide [see Figure
2(b)].
EXAMPLE 22 Find the solution of
Ix + y = 3 (1)
6x + 2y = 12 (2)
SOLUTION Multiplying the first equation by -2 gives
-6x - 2y = -6 (1)
6x + 2y = 12 (2)
0=6
Adding equations 1 and 2 yields on the left side of the equation and 6 on the
right.Since + 6, no numbers x and y satisfy both equations. Solving for y in
the two equations gives
y = -3x + 3 (1)
H-D-f
The slopes of the two lines are equal and the y intercepts are different, so the
lines are parallel. The solution to the system is the null set.
Let us now examine a calculator approach to solving a system of

equations.
EXAMPLE 23 Solve
2.33* + 3.01y = 7.72 (1)
3.71a: - 4.32y = 3.09 (2)
using a calculator.
SOLUTION Suppose we decide to eliminate x. Then, with a calculator, we

multiply thefirst equation by 3.71 and the second equation by 2.33. This
makes the coefficient of x the same in both equations. We then subtract:

[3.71(3.01) - 2.33(-4.32)]y = 3.71(7.72) - 2.33(3.09)
y = 3.71(7.72) - 2.33(3.09)
3.71(3.01) - 2.33(-4.32)
CALCULATOR NOTE Can you perform these calculator operations?
3.71 H Q7.72 2.33 3.09 [7] H 3-71 3-01 DE H2-33
4.32 SHU] ED y= 1.0098
Substitution of y = 1 .0098 in the first equation yields
2.33.x + 3.01(1.0098) = 7.72

v = 2.0088
Check: 2.33(2.0088) + 3.01(1.0098) = 7.72

3.71(2.0088) - 4.32(1.0098) = 3.09
The answer gives accurate results to two decimal places. The solution is
(2.0088, 1.0098).
The following illustrations are examples of the use of systems of two

linear equations in two unknowns in solving practical problems.
EXAMPLE 24 The total number of Democrats and Republicans in a community is 40,000.

In a recent election, 60% of the Democrats and 40% of the Republicans
voted. If only Democrats and Republicans voted and the total vote was
21,400, find the number of Democrats and the number of Republicans in the
community.
SOLUTION Let's apply the problem-solving techniques of Chapter 1 as a start

toward the solution to this problem. First we will guess an answer and then
check with a calculator in Table 2. Then we replace our guesses with vari-
ables x and y.
TABLE 2
Break-Even Point
When an equation expressing total cost C in terms of * (the number of items
produced) and an equation expressing revenue R in terms of * are graphed
on the same coordinate system, then the break-even point is the intersection
of the two lines. Of course, the break-even point can be found by solving the
system of linear equations by setting C = R.
EXAMPLE 25 A university is offering a special course in crafts for which tuition is $60 per
student. The university has found that the cost of the course is $600 plus $20
for each student who registers for the course. How many students must take
the course for the university to break even?
SOLUTION Let x = the number of students taking the course. Then the reve-
nue equation becomes R = 60*, and the cost equation becomes C = 20* +
600. Setting these two equations equal to each other gives
R = C $600 is the fixed cost |
$60* = $20* + 600 $20* is the variable cost
40* = 600
* = 15
The graph of these two functions is given in Figure 3. The solution (15, $900)
means that the university receives $900 if 15 students take the course. Like-
wise, $900 is the cost for 15 students.
Practice Problems 1. Solve the system of equations
3* + y = 5
* = 2y - 3
Price
Hint: You need to arrange the variables in the same order.
2. Solve
3jc - 2y = 4
Ay - 1 = 6x
Exercise Set 2.5 . Solve

5. The sum of 2 test scores is 175. (d) 137.v - 41 v = -10,688

The difference between the two 105* + 13y = -2,994
scores is 11. Find the scores.
Exam. Given the following
notations, what is the break-
6. Fifty coins totaling $7.40 are re-
even sales level in units? Hint:
moved from a ma-
soft-drink
chine. If the coins are dimes
all
C = FC + (VCH.v), and R =
(SP)(.\). where x is the sales
and quarters, determine the
level, SP = selling price per
number of each.
unit, FC = total fixed cost, and
VC = variable cost per unit.
7. Solve the following systems by
using a calculator: SP
(a)
FC - VC
(a) 17.05* - 3.24y = 22.63
FC
3.21.V - 4.56y = 3.96 (b)
VC -h SP
(b) 35.7.V + 103v = 104.759 VC
(c)
-23.4a: + 37y = 118.412 SP - FC
(c) 123.v - 37;y = 6890 FC
(d)
47.v - 3ly = 896 SP - VC
Applications
(Business and 9. Mixture Problem. A candy- 11. Break-Even Point. A producer
Economics) store proprietor wishes to mix knows that shecan sell as many
candy that sells for $3 per pound items at $0.25 each as she can
with candy selling for $4 per produce in a day. If her cost is
pound to make a mixture to sell C = $0.20jc + $70, find her

for $3.60 per pound. How many break-even point.
pounds of each kind of candy
12. Break-Even Point. A firm knows
should be used to make 80
that can sell as many items at
it
pounds of the mixture?
$1.25 each as it can produce in a
10. Commission. Jim is trying to de- day. If the cost is C = $0.90* +
cide between two positions. The $105, find the break-even point.
first pays $225 per week plus 5% 13. Break-Even Point. If the firm in
commission on gross sales. The
Exercise 12 can change the cost
second pays only 9% on gross = +
equation to C(x) $0.80.v
sales. Graph the two pay func-
$120, should the change be
tions and find where they are
made? Explain.
equivalent.
and Life Sciences) 14. Nutrition. A special diet re- traction and repulsion by first
quires 4 milligrams of iron and feeding mice and then later giv-
52 grams of protein each day. A ing them mild electric shocks
person decides to attain these re- from the same box. With this
quirements by drinking skim procedure the psychologist es-
milk and eating fish. A glass of tablished the following functions
skim milk provides 0.2 milligram where a represents attraction, r
of iron and gram of protein.
1 represents repulsion, and x rep-
One-fourth pound of fish pro- resents the distance in centi-
vides 8 milligrams of iron and 10 meters of the mouse from the
grams of protein. How many box.
glasses of milk and how many
a = - -. x + 70
pieces of fish (| pound) are 4
needed to attain the diet's re-
4
quirements? r = - - x + 200
15. Population. A town has a popu-
Graph the attraction function
lation of 1000. The number of
and the repulsion function on the
men is 80 less than twice the
same coordinate axes, and find
number of women. Find the
the distance where attraction
number of men and the number
equals repulsion. (Hint: Find
of women.
the intersection point of the two
16. Psychological Attraction and Re- lines.) Check your graphical
pulsion. A psychologist has result by setting a = r and solv-
been studying reactions of at- ing for x.
Answers to 1. The solution is (1, 2).
Practice Problems 2. The solution is the null set because the two lines are parallel.
2.4 Solving Systems of Linear Equations with Matrices
Overview In practical applications, most systems of linear equations involve a larger number of equations
and unknowns. Usually these systems are solved using computers. In this section we introduce
a procedure for solving systems of equations that can be extended to any number of equations
and unknowns.
Our first step will involve forming what we call an augmented matrix. We then perform
permissible matrix operations to put the matrix in a form where the solution is obvious. The
step-by-step procedure we use is called the Gauss-Jordan elimination method.
An augmented matrix for a system of equations contains the coefficients of

the unknowns as elements, in all except the last column. The elements are
listed in the same order as they appear in the system of equations. A vertical
Solving Systems of Linear Equations with Matrices 93
line usually replaces the equal signs. The last column consists of the con-
stant terms. For example, the augmented matrix for the system
2x + 3y = -5
x - 2y = 8
2 3 -5
1 -2
EXAMPLE 26 Find the augmented matrix for the system
2x = 7 + 3y
4y = -4 - 3.v
SOLUTION First we write the two equations with the variables occurring in
the same order.
2x - 3y = 7
3x + Ay = -4
The augmented matrix is
2 -3 7
3 4 -4
EXAMPLE 27 Using x, y, and z, write a system of linear equations having the augmented
matrix
From the preceding example it seems that if there is a set of operations

which we can perform on an augmented matrix that will reduce it to the form
shown in Example 28, this may give a procedure for solving a system of
equations.
Now let us demonstrate three operations on rows of a matrix that give
a new
matrix representing a system of equations with the same solution as
the original system, where /?/ represents the th row. (For example, Z? 3 is the i
third row.)
Row Operations 1. Interchange two rows (fl, <- fly, or interchange the /'th row and the/'th row).
2. Multiply (or divide) each element of a row by a nonzero constant (cfl, - fl,, or replace
the /'th row by a constant times the /'th row).
3. Replace any row by the sum (or difference) of that row and any other row times a
constant (can be 1 ) or replace the/th row by a constant times the /'th row added to the
/'th row (cfl, + /?,)-/?,-.
1. Interchange two rows. For the matrix
3 -
interchange the rows (/?, -> R 2 ):
3 -
2
The matrix
2
3 -
represents
2x + y = 3
3* - y = 2
while
3 -1
2 1
represents
3* - y = 2
2;c + y = 3
You can verify that x= 1, y = 1 satisfies both systems of equations. Thus,

interchanging rows does not affect the solution.
Multiply (or divide) each element of a row by a nonzero constant. For

the
2.
matrix
3 2
1 -7
multiply the first row by 4 (4#, -* fl,). The new matrix is
These represent the systems of equations

3x + 2y = 1
2x - 2>> = 4
and
12* + 8y = 4
2x - 2y = 4
Verify that these two systems have the same solution, namely, x 1.
y= -l.
3. Replace any row by the sum (or difference) of that row and any other row
times a constant. Using the matrix
4 1
3 2
let's replace the second row with the sum of the second row and 2 times
the first row, or 2/?, + R2 ~ ^2- Replace 3 with 2

4 + 3 = 11,2 with 4,
and with 10, getting
4 1
11 4
Show that (2, -3) is a solution of both
4.v + y = 5
3.v + 2y =
and
4* + v = 5
ll.v + 4y = 10
Thus, the row operations as defined produce systems with the same
solution. If by row operations we can get an augmented matrix in the form
1
1
we will have the solution
y = b
and this satisfies the original system of equations. The procedure for trans-
forming an augmented matrix to this form is called Gauss-Jordan elimina-
tion after Carl Friedrich Gauss (1777-1855) and Camille Jordan (1838-1922).
EXAMPLE 29 Solve
2x + 3y = -5
x - 2y = 8
by augmented matrices.
SOLUTION The augmented matrix is
2 3 -5
1 -2
In the first two steps we will be working to obtain a matrix that looks like
1 e e, f, g, and h can
g be any real numbers
(Get the 1 first and then the 0.) In the last two steps, we obtain the second
column to look like
1 j and k can be
1 any real numbers
(We will follow the procedure of getting the 1 first and then the 0.)
NOTE We could get a in the upper left corner of this example by interchanging the
1
firstand second rows. Likewise, we could solve this small system more
easily using something other than augmented matrices. However, on this
small system we will practice procedures that will, in general, work on more
complicated problems.
Need a
1 here
Divide each element of the first row by 2:
i*
Need a
here
Now multiply the first row by - 1 and add to the second row:
-fl, + R : -* R 2
Multiply the second row by -f
= R 2 -> R2
Need a 1 2
1 2
here 1
Multiply the second row by -f and add to the first row:
3
R-, + R, R,
Thus,
The solution is (2, -3).
EXAMPLE 30 Solve
using augmented matrices.

1 2
1 3
This is the augmented matrix for the system
1 5
" V + y
=
2 2
x +
y = 2
and because the second equation has no solution, the system has no solu-
tion.
EXAMPLE 33 Solve the system of equations

3jc - 2y = 1
6.v - 4y = 2
by using augmented matrices.
SOLUTION The augmented matrix of the system can be written as
*/?,-/?
1
1
- 23
Need a
6 -4 + R^
-6R,
T
1
1
23
The system of equations can be written as
2 1
*~3-V =
3
x + -y =
Since any pair of values that satisfies the first equation will obviously satisfy
the second, there are infinitely many solutions. For example, let y = /. Then
Summary of Solutions
EXAMPLE 34 Pepident Toothpaste Company manufactures two kinds of toothpaste in

quantities measured in units equivalent to 1000 5-ounce tubes. Some of the
toothpaste is sold to discount stores under several house brands at a net
profit of $20 per unit; toothpaste sold under Pepident's own brand nets $30
profit per unit. If sales last month were 50,000 units and the profit was
$1,300,000, how many units were sold as "Pepident" and how many under
another brand?
SOLUTION Using our problem-solving techniques, we first make a guess and

then replace our guess with variables.
Number of Units of Pepident Other Brands Total
Guess 20,000 + 30,000 = 50,000

Profit 20,OOO($3O) + 30,000($20) = $1,200,000
(too small)
Assignment x + y = 50,000
of Variables
Profit 30x + 20y = 1,300,000
The augmented matrix of the system is
[1 1 j 50,0001
L30 20 1,300,000J
2. Write a system of equations from the augmented matrix
1 3-1 -2
4 -1 3
3 -2 -4 -1
3. Write a system of equations from the augmented matrix
4. Using augmented matrices, solve the system of equations
x + 3y = 5
x + 5y = 7
Exercise Set 2.4 Write the system of equations for

each of the following augmented
matrices:
I.
1 4 Discuss the solutions for the follow-

(c)
ing systems of equations:
Solve each of the following systems 21. Ax + 3y = 7

of equations by using augmented 8.v + 6y = 14
matrices:
22. 2.v 4y = 6
13. Exercise 1 3x 6v = 9
14. Exercise 2 23. 2x - y = 3

Ax - 1y = 9
15. Exercise 5
24. 2x + 4y = 1
16. Exercise 6
1x + 6y = 11
17. Exercise 7
Use a calculator to solve the follow-
18. Exercise 8 ing systems of equations:
19. Exercise 9 25. I.Ijc + 2.5y = 1

l.lx - 3.5y =6
20. Exercise 10
26. O.Ijc + 0.6>> = 7
I.Ijc + 1.6>> = 4
Applications
(Business and 27. Production Scheduling. In a 29. Mixture Problem. The Reused
Economics) small furniture-manufacturing Paper Company uses both scrap
plant, 400 hours of labor are paper and scrap cloth to make
available for making tables and their paper. Their best paper re-
chairs and 107 hours are avail- quires 3 tons of cloth and 15 tons
able for finishing (painting or of paper for each run, but their
staining). If it takes 8 hours to good paper requires 1 ton of
make a table and 5 hours to cloth and 12 tons of paper for
make a chair, and if it takes 2 each run. How many runs of
hours to finish a table and 3 their best paper and how many
hours to finish a chair, how runs of their good paper should
many of each can be manufac- be made if Reused Paper has 34
tured using this schedule? tons of scrap cloth and 261 tons
of scrap paper on hand? Assume
28. Investments. The Easy Invest-
Reused Paper wishes to use all
ment Club has $200,000 invested
its scrap paper and scrap cloth.
in bonds. Type A bonds pay 8%
interest, and type B bonds pay 30. Mixture Problem. At a recent

10%. How much money is in- concert, were $5 for
tickets
vested in each type if the club adults and $3 for students. There
receives $18,720 in interest? were 2000 more adult tickets
Systems with Three or More Variables or Three or More Equations 103
sold than student tickets. How 31. Break-Even Point. A revenue

many tickets of each kind were equation is given by R = 1.40*
sold if the total receipts were + 60, where x > 10 is the num-
$42,000? ber of items produced. The cost
equation is C = 0.95* 4- 105.
Find the break-even point.
and Life Sciences) 32. Nutrition. A dietician plans to drate per unit and food II with 4
combine food A and food B to grams of carbohydrate per unit.
make a meal containing 2000 cal- To keep him from becoming dis-
ories, and 30 units of a combina- couraged with his diet, Abdul's
tion of vitamins. Each ounce of doctor has insisted that he con-
food A contains 200 calories and sume 500 calories at each meal,
5 units of vitamins. Each ounce of which 210 calories must be
of food B contains 250 calories protein. Both foods contain 100
and 4 units of vitamins. How calories per unit, but food I con-
many ounces of each food tains only 30 protein calories per
should be in the meal? unit, whereas food II contains 50

protein calories per unit. Help
33. Diet. Abdul is on a low-carbo-
Abdul decide how many units of
hydrate diet. He is planning a
each food he should eat to con-
meal composed of two foods:
sume the calories allowed.
food I with 7 grams of carbohy-
Answers to t>
Practice Problems l.

m = n : The number of equations is the same as the number of unknowns.

m> n: There are more equations than unknowns.
m< n: There are more unknowns than equations.
Systems with a large number of equations and variables are very common today because of the
accessibility of high-speed computers.
The solution of an equation in three variables, such as
jc + 2v-3z + 6 =
is an ordered triplet (x, y, z). For example, (-3, 0, 1) is a solution since
l(-3) + 2(0) - 3(1) + 6 =

-3-3+6=0
The solutions for a system of three equations in three unknowns are the
ordered triplets that satisfy the three equations.
The augmented matrix for three equations with three variables is very
similar to that for two variables (has one additional column and one addi-
tional row).
EXAMPLE 35 Find the augmented matrix for the system
SOLUTION
SOLUTION
EXAMPLE 37 Solve the system of equations
x + 2y - 3z = -6
2x - y + z = -1
3.v + 2y + z = 4
SOLUTION The augmented matrix for this system is
2 -
Each of these
needs to be
This needs
to be a 1
solution The augmented matrix is
0.2/?,+ /?, - Rt
4R, + R,-^ R,
The solution is x = -1, y = 3, and z = 4.
Linear equations in three variables represent planes in three-dimen-

sional space; visualizing how you insight as to
three planes intersect can give
what type of solutions are possible with a system of planes (Figure 4).
Sometimes it is not possible to reduce a matrix of coefficients to an
identity matrix. You will note that only one graph in Figure 4 has a unique
point on all three planes (case 1). This case corresponds to reducing the left
side of an augmented matrix to the identity. All other graphs correspond to
cases where the matrix of coefficients does not reduce to the identity matrix.
Thus, we need to define what we mean by a matrix in reduced form.
FIGURE 4
Description Graph
1. The common intersection consists

of a single point.
2. The intersection is a line, and the

system has infinitely many solu-
3. The three planes coincide, and the

system has infinitely many solu-
tions.
4. The three planes are parallel and

have no common intersection, and
the solution set is the null set.
(continued)
FIGURE 4 (continued)
Description Graph
5. Two planes are parallel, and thus

the solution set is the null set.
6. The planes intersect only in pairs

of parallel lines, and the solution is
the null set.
Definition: Reduced A matrix is in reduced form if:
Matrix
1. The leftmost nonzero element in each row is 1.
2. The leftmost nonzero element in a row has all O's above it and all O's below it in its
column.
3. The first nonzero element in each row is to the right of the first nonzero element in
each row above it.
4. Rows containing all O's are below the rows containing nonzero elements.
Suppose the last equivalent matrix of a system is of the form
1
The value of c, which usually comes from the last equation, can be anything
and the last equation is satisfied (case 2, Figure 4). Thus, we say the value of
Z is arbitrary, or we can assign any real number as the value of z. For
example, let's assign z = 3. Then
1
x + y + 6 = -3
x + 1 y - 6 = 4
0-JC + 0-y + 0-3=

In the first equation x = -9. and in the second equation y = 10. One solution
is (-9, 10, 3).
Suppose we let z be an arbitrary constant c. Then
y 2c = 4 or y = 4 + 2c
x + 2c = -3 or x = -3 - 2c
Then .v, v, and z are all expressed in terms of a constant c. The solution can
be expressed as
(-3 - 2c, 4 + 2c, c)
for any real number c. When c = 1, the solution is (-5, 6, 1). When c = 2, the
solution is (-7, 8, 2).
EXAMPLE 39 Discuss the solution of the system of equations when the reduced augmented
matrix is of the form
1
EXAMPLE 40 Solve the following system by operations on the corresponding augmented

matrix.
ly + 3z - It = 3
2x + z - 2/ = -5
4x + y + z - 5? = -6
y + 2z - At = 1
SOLUTION The augmented matrix is
The reduced matrix is
Since the bottom row of the matrix consists entirely of 0's, the value of t is
arbitrary. Let c be a number chosen in any fashion but left fixed for a
moment; then let
Systems with Three or More Variables or Three or More Equations 1 1
3. Find the solution of the system of equations with variables x, y, and z and
augmented matrix
I 2
1 3
4. Find the solution of the system of equations with augmented matrix
1 2
1 3
5. Write a system of equations in terms of x, y, and z for the augmented

matrix
[1 2 3
LO 1 1 2
and discuss the solution of x and y.
Answers to practice problems are found on page 1 14.
Exercise Set 2.5 Indicate whether each matrix is in

reduced form.
ri o
Li o
13.
14.
15.
16.
17.
18.
x - v + z = 2 (a)
(b) l
2x - y + 2z = 2
(c) 2
3.v + 3v + 3z = 4 (d) 3
(e) x
Applications
(Business and 33. Investments. Jeannette has utes on station I, 5 minutes on
Economics) $9000 invested in bonds paying station II, and 15 minutes on sta-
14%, 16%, and 18% interest per tion III. Each B requires 15, 10,
year. She has $1000 more in- and 5 minutes on I, II, and III,
vested at 16% than at 18%. If respectively; each C requires 20,
she receives $1430 interest each 15, and 5 minutes, respectively;
year, how much money does she and D requires 5,
and 15 min- 10,
have invested in each type of utes, Find the
respectively.
bond? number of batteries of each kind
that can be manufactured in a
34. Investments. Sarah has a total
day if station I is available for
of $12,400 invested in three
740 minutes, station II for 480
types of bonds pay 9%, that
minutes, and station III for 660
10%, and 11%, respectively. She
minutes.
receives $100 more interest from
the 10% bonds than from the 36. Demand Curve. The economist
11% bonds. If the total income at Acme
Corporation has noted
from the bonds during a year is that the demand curve for a
$1250, how much has she in- given product is in the form of a
vested in each type? parabola, p = ax2 + bx + c. The
demand price for 2 units is $1;
35. Manufacture. Safety Battery
that is, p = 1 when x = 2; p = 2
Company manufacturers 4 kinds
when x = and p = \ when x =
1 ;
of batteries, grades A, B, C, and

3. x
Substitute these values for
D. Each battery must go through
and p and solve the three linear
3 process stations in manufac-
equations for a, b, and c. Predict
ture. Each A requires 10 min-
demand price when x
the is 4.
and Life Sciences) 37. Pollution. The pollution count 38. Truffle Flow. The number of
for Clanton on a particular day is cars entering and leaving 4 inter-
600. Assume that this pollution sections of one-way streets has
isproduced by 3 industries: A, been tabulated as shown in the
B, and C. Industry A contrib- diagram. The number of cars en-
utes twice as much to the pollu- tering an intersection must equal
tion count as industry B. It is the number that leave. At B, 250
known that the pollution count + x 4 = 150 + xi Find the equa- .
would be 500 if the pollution tions at C, D, and A. Set up and

count from industry A were re- find a reduced augmented ma-
duced by 50%. Find the pollu- trix. Note that you can solve for
tion count of industries A, B, the three variables in terms of
and C, respectively. x4 What
. is the smallest value x 4
Inverse of a Matrix and Its Use 115
are inverse matrices since

r
Definition: Inverse
of a Matrix
solution To find the inverse of
define the inverse to be
b d
1
AA-
ll 2J lb d\ 1
2a + 3b 2c + 3d
a + 2b c + 2d 1 = P
Lo
1
lJ
So
2a + 3b = 1 and 2c + 3d =
a + 2b = c + 2d = 1
Solving these two systems of equations, we find a = 2, b = 1, and c = -3,

d = 2. Thus, the inverse is
r 2 -31
Check:
which can be proven by noting that
5
This fact can be proven by multiplying:
2 r
where A is the coefficient matrix
and
X =
The inverse of
SOLUTION In matrix notation this system becomes
The inverse of
4. Find the solution of the system
2x + 3y = 7
3.v + y = 1
by finding the inverse of the coefficient matrix. t>
Answers to practice problems are found on peine 125.
Exercise Set 2.6 1. Show that the following matri-

ces are inverses of each other:
3 -2
(a)
4
(b)
(c)
(d)
Applications
(Business and 17. Investments. The Lake shore 19. Break-Even Analysis. A country
Economics) Investment Club has $100,000 club has been studying its cost of
invested in bonds. Type A bonds operation and has found the
pay 8% interest, whereas type B monthly cost to be C = $6000 +
bonds pay 10%. How much $15x, where x represents the
money is invested in each type number of members. A service
if the club receives $9360 in company states that by remodel-
interest? ing, it can provide a similar ser-
vice for a monthly cost of C =
18. Investments. Jane has money
$10,500 + $10*. Use the inverse
invested through Merrill Lynch
of a matrix to find the break-
at 12% annual interest, a bank
even point, and find the number
investment certificate that pays
of members necessary to under-
10% interest, and a personal
take the remodeling.
loan to a friend at 5% interest.
She remembers that she has 20. Manufacture. A company makes
$16,000 invested and that the 2 kinds of tires. Type I requires
amount with Merrill Lynch is 2 hours on machine A and hour 1
$4000 more than that invested at on machine B. Type II requires

the bank and loaned to the 3 hours on A and 2 on B. How
friend. Her annual interest is many tires can be manufactured
$1750. Use the inverse of a main a week if machine A is in use
trix to help Jane determine how 40 hours and machine B 25

much she has invested in each hours?
account.
and Life Sciences) 21. Herbicides. Three herbicides gallons of each herbicide should
are available to weeds,
kill be purchased?
grass, and vines. One gallon of
22. Diet. A dietician wants to com-
one herbicide contains 1 unit of
bine 2 foods as the main part of a
chemical A, which kills weeds,
meal to get 2600 calories and 70
and 2 units of chemical B, which
units of vitamin C. Each ounce
kills grass. One gallon of the sec-
of food I contains 200 calories
ond herbicide contains 2 units of
and 10 units of vitamin C. Each
chemical B and 3 units of chemi-
ounce of food II contains 400
cal C, which kills vines. One gal-
calories and 5 units of vitamin C.
lon of the third herbicide con-
How many ounces of each food
tains 3 units of chemical A and 4
would be required to obtain the
units of chemical C. It is desired
desired number of calories and
to spread 10 units of chemical A,
the desired units of vitamin C?
16 units of chemical B, and 20
units of chemical C. How many
Input-Output Analysis 125
23. Polling. A company has been whereas team II can make 70

engaged to make 4850 phone phone polls and 30 home polls
polls and 2650 home visits. The each day. How many days
company has 2 teams of poll- should each team be scheduled
sters. Team I can make 60 phone in order to complete the engage-
polls and 40 home polls a day. ment?
Answers to [-2 1
Practice Problems
1. A-'A =
AA
any industry, an is the fractional amount of item 1 needed by industry 2 to

produce one unit of item 2. a 2 i is the fractional amount of item 2 needed by
industry 3 to produce item 3. In general, a v is the fractional amount of the /th
item required as input by industry j to produce one unit of the jth item as
output. Written as a matrix, this becomes
1 1
X] = = X] + r
2
1
-v: = t -vi
In matrix notation this system can be written
X 2.
Note in the sum of the elements in each

preceding example that the
column is For a closed Leontief system, the sum of each column of the
1.
Leontief input-output matrix must equal (unity) because all of the output is 1
consumed as input.
The expression ai\X\ + 0,2*2 +
+ a m xn represents the total input
requirements of industry i of all the items. For a closed Leontief system, the
number x, of output units of item i by industry /'
must equal the total input
requirements by industry /'
of all the items. Hence,
x\ = aux t
a i2 x 2
x2 = a 2] Xi a 2 ->x-> + a.
x = a\X l
+ a 2 x 2 + + a nn x
In matrix notation this becomes X = AX, where
flu
be a column matrix representing the dollar amount paid to each by the others
(italso includes the dollar amount one individual would pay to himself; for
example, the farmer pays himself for the food he consumes). Since the total
amount paid out by each must equal the total amount received by each one.
X\
represents the number of units of item demanded by the consumer.

and di
i
Instead of
K1U] = [x,] or AX = X I
Closed system |
we now have
la.Mx,} + [di] = M
where A = [a,j] and X = [.v,] are denned as before and d
represents the t
number of units of item i demanded by the consumer. If D = [</,], then

AX + D = X
D = X - AX Subtract AX from both sides
D X= - output available to the consumer sector of the

AX represents the
economy. A very fundamental problem is whether or not the economy can
satisfy the consumer demand. Another way of stating this
problem is: Given
matrix X with
a demand matrix D of the consumer sector, does there exist a
nonnegative components such that X - = D? This question will be AX
answered after the next example.
EXAMPLE 49 To illustrate an open Leontief system, let us return to Rural City and assume
that the farmer and the tailor decide to sell some of the items they produce to
a consumer. Suppose the farmer requires one-third of the food
and one-third
tailor requires
of the clothes to produce one unit of the food. Suppose the
unit of the
one-half of the clothes and one-fourth of the food to produce one
clothes. Write these requirements as an open Leontief
input-output matrix.
SOLUTION
Food Clothes
produced produced
Input Farmer
Requirements Tailor
This matrix indicates that in order for the farmer to produce food, he
requires one-third of the food produced and one-third of the clothes pro-
duced. Similarly, in order for the tailor to produce clothes, he requires one-
fourth of the food produced and one-half of the clothes produced. Let x,
represent the number of units of food produced and x 2 represent the number
of units of clothes produced. Suppose d represents the units of food and d 2
x
the units of clothes required by the consumer. The requirements

can be
written
3 4
We are now ready to return to the fundamental question raised con-

cerning the open Leontief system. Remember that the question was: Given a
demand matrix D of the consumer sector, does there exist a matrix X such
that
X - AX = D
If I is the identity matrix, this equation becomes
IX - AX = D or (I - A)X = D
If (I - A) _i exists, then
X = (I - A) 'D
Hence, if (I - A) 'D exists and has nonnegative components, it provides an

affirmative answer to the fundamental question.
EXAMPLE 50 In Example 49, AX + D = X was
l 1
3 4
1 1
3 2
EXAMPLE 51 A three-industry open Leontief system has the following input-coefficient

matrix
Services
Exercise Set 2.7 Find one solution for each of the

following systems.
0.7
12. The input-output matrix of a quirements as a Leontief input-

closed Leontief system is output matrix A. If
"HO"
111
gross production is required to

meet an external demand for 50
units of electricity, 60 units of
oil, and 30 units of coal?
Answer to > 1 . For one unit of production:

(a) A consumes 0.3 of what it produces.
(b) A consumes 0.6 of what C produces.
(c) B consumes 0.2 of what A produces.
(d) C consumes 0.2 of what B produces.
Graph Theory and Matrices
Assigning broadcast frequencies, scheduling work assignments, and
making production runs at a factory are just a few examples of the uses of
graph theory. In this discussion we will show how matrices are useful in
representing graphs. In graph theory, a dot represents a vertex, and a line or
a curve connecting two dots is called an edge. For example, A, B, and C are
vertices, and an edge connects A and B, B and C, but not A and C. Some-
times graphs are used to represent relations between elements of a set. In
this case, there could be a relation from a vertex to itself. Such an edge is
called a loop. A loop appears at V, below.
(1)
(2)
A representation of a graph can be put into a computer using a matrix. One

way is to define the elements of the matrix as follows.
_ 1 1 if there is an edge joining V, and V,

,J
lo if there is no edge joining V, and V,
Extended Application 135
Matrices representing the graphs above are:
ABC
A
V, (CPU) 7
V 2 (Memory)
l , (Input) I
4 (Output)
The matrix for this graph would be
1
Product of matrices (74)

Dot product (73)
Multiplicative inverse (1 15)
A" where A is a square matrix (76)
System of Two Linear Equations

Graphical solutions (82)
Algebraic solutions (85)
Two intersecting lines (82)
Addition and subtraction (85)
Two parallel lines (84)
Break-even analysis (89)
Coincident lines (84)
System of Three or More Linear Equations

Geometric interpretation (107)
More equations than unknowns (104)
More unknowns than equations (104)
Augmented Matrices
Permissible row operations (94)
Reduced matrix (108)
Gauss-Jordan elimination (96)
Solutions of systems of equations (96)
Inverse of a Matrix
Solution of systems of equations (120)
Input-output analysis (130)
Review Exercises 1. Determine whether or not the 2. Solve the following systems of
following systems have common
solutions, and solve by graphing
and by the addition-subtraction
method:
(a) 3jc + 2y = 1
5x - 3y = 27
(b)
3. Form matrices representing the 5. Construct a directed graph for

following graphs: the following matrices.
(a) v, n (a)
(b)
(b) v
Chapter Test 139
8. Solve the following systems of 11. Discuss the solutions of the fol-
equations by using augmented lowing:

matrices. (a) 3x + 2y = 5
(a) x + y + 2z = 4 Ax - y = 3
3x - y + z = 3 x + y = 2
5x + 3v - 4z = 4 (b) 2x - y =
(b) 3jc - y - z = 2 3x + 2y = 8
a: + 4j - 3z = 11 -x + 2y =
-v - y + 2z = 1 (c) 5* + ly + z = 13
2x - y + z = 2
9. Solve the systems of equations
(d) 6x + 5y + z = 17
in Exercise 8 by using inverses
x - 2y + z = -2
of matrices.
12. Determine whether or not the
1(1. Discuss the solutions of the fol-
following systems have solu-
lowing.
tions, and solve those that do:
(a) (a) X +
(b)
(b)
(c)
(d)
(e) 1
(f) 10
I
1
2 3"
3. 3
Introduction to Linear
Programming
We all like to get as much as possible

exchange for the least amount of
in
effort or expenditure. Remember the time you went shopping, you
last
wanted to get the most for your money. And have you ever tried to satisfy
your appetite with only 1000 calories a day? All of us have struggled with our
daily schedule, trying to find enough time for classes, for study, for a part-
time job, and still have the maximum amount of leisure time for club activi-
ties, movies, dates, and other activities.
Of course, the businessperson, the manufacturer, the planner, as well

as the researcher, are just as interested in minimizing expenditures and
maximizing benefits as you are. Thus, the mathematical model we are about
to study, linear programming, is a valuable tool.
This application of mathematics is rather young compared with other
mathematical ideas. During World War II, the Allies faced daily problems
concerning the transportation of troops, the movement of supplies, and the
scheduling of operations. Linear programming models helped to solve these
problems. George Dantzig (1914- ) is given credit for developing the com-
puter-oriented approach to solving linear programming problems, a proce-
dure called the simplex method. For his pioneering contributions in this field
and the development of the simplex procedure, in 1947 Dantzig was awarded
the National Medal of Science.
We begin this chapter by locating graphically the solutions of a system
of inequalities. This study of a system of inequalities helps us to solve
141
142 Chapter 3 Introduction to Linear Programming
geometrically linear programming problems. We then study a procedure for

solving linear programming problems that can be used for large computer-
oriented problems.
^ 3.1 Solving Systems of Inequalities Graphically
Overview Mathematical models often contain inequalities that express limitations. For example, the size
of a school building limits the number of students who may attend the school; the amount of
money a person has limits the amount he or she can invest; the quality of steel a company
wishes to produce places conditions on the ingredients; and so on. In this section we will:
Solve systems of inequalities graphically

Formulate special definitions for concepts associated with a region of solutions for a system
of inequalities
Apply systems of inequalities
1, we studied the graphs of linear equations and linear

In Chapter
inequalities. In Chapter 2 we studied the graphs of a system of linear equa-
tions. In this section, we study the graphs of a set of linear inequalities called
a system of inequalities. The solution region for a system of inequalities in
two unknowns consists of the ordered pairs that satisfy all the inequalities.
EXAMPLE 1 Graph the solution region of the system
2x + y - 1 >
x - y + 1 <
SOLUTION Solutions of linear inequalities were graphed in Chapter 1 . In Fig-

ure 1 2x + y - 1 > 0, and
the light-gray shading represents the solutions of
the lightly colored shading represents the solutions of jc y + 1 ^ 0. These
two solution regions are also indicated by two arrows perpendicular to the
boundary of each inequality. The doubly shaded region represents the solu-
tion region, or solution set, of the system, consisting of all ordered pairs
satisfying both inequalities.
EXAMPLE 2 Solve the following linear system graphically:
x >
x < 4
y >
y<3
Solving Systems of Inequalities Graphically 143
Zx + v - l =
FIGURE 1
SOLUTION The graphs of the four inequalities are shown in Figure 2. The
solution region associated with each line is indicated by arrows. The inter-
section of the four graphs (or the solutions common to the four graphs) is the
solution region of the system and is shaded>
FIGURE 2
Note in this definition that the intersection of two boundary lines may
or may not be a corner point. The intersection must be in the solution region.
In Example 1, (0, 1) is a corner point. There are four corner points in
Example 2: (0, 0), (0, 3), (4, 0), and (4, 3).
EXAMPLE 3 Graph the solution region and find the corner points of the system
2x + y < 3
-x + y <
x >
y >
SOLUTION The solution regions for the inequalities 2x + y < 3, x + y ^ 0,

x > 0, and y > are graphed in Figure 3, and the solution region of the points
satisfying all the inequalities is shaded. The corner points for this system are
A = (1, 1), B= (0, 0), and C= (J, 0), as illustrated in Figure 3. Why is (0, 3)
not a corner point?^
FIGURE 3
Solution regions may be bounded or unbounded.
Definition: Bounded A solution region of a system of linear inequalities is said to be bounded if one can draw a
Solution Region circle that will contain all points of the solution region. If the solution region cannot be
contained in some circle, it is said to be unbounded.
The solution regions of Figure 2 and 3 are bounded regions; however,

the solution region of Figure 1 is unbounded.
EXAMPLE 4 Find the corner points and graph the solution region of the system
2x + y < 3
-x + y <
x <
FIGURE 4
SOLUTION The lines 2x + y = 3, -x + y = 0, and x = are graphed, and then

the solution set of the points satisfying all is shaded in
of the inequalities
Figure 4. The only corner point is B = (0, 0). The
and (1, 1) are
points (0, 3)
intersections of boundary lines but are not corner points because they are
not part of the solution region of the system. Note that this solution region is
unbounded.
In the first of the two applications of a system of inequalities, we will

use the problem-solving ideas detailed in Chapter 1. If you need these aids,
do not hesitate to use them on Example 6.
EXAMPLE 5 A supermarket mixes its 60% beef and 40% fat. For
regular ground beef as
extra-lean ground beef, the market uses 75%
beef and 25% fat. The store has
up to 225 pounds of beef and up to 125 pounds of fat available to make
ground beef. Write the system of inequalities that expresses these condi-
tions, graph the solution region, and find the corner points.
solution Suppose you guess that the store wants 200 pounds of regular
ground beef and 40 pounds of extra-lean beef. First, we determine whether
we can get this guess from 225 pounds of beef and 125 pounds of fat. The
regular ground beef is 60% beef, so we need
.60(200) pounds of beef |

Our guess
The lean beef is 75% beef, so we need
.75(40) pounds of beef. |
Our second guess
Since we have Only 225 pounds of beef,
.60(200) + .75(40) < 225

Verify with a
calculator that this
inequality is satisfied.
Regular ground beef is 40% fat. So
.40(200) pounds of fat are needed From our guess

of 200 pounds.
The lean ground beef is 25% fat. So
.25(40) pounds of fat are needed From our guess

of 40 pounds.
Since we have only 125 lbs. of fat,
.40(200) + .25(40) 125
If both of the above inequalities are satisfied, we have indeed guessed a

solution. However, we want all solutions, so we replace 200 with x (repre-
senting the number of pounds of regular ground beef) and 40 with y (the
number of pounds of extra lean beef) to obtain
0.60* + 0.75y < 225 Condition on beef
In a similar manner,
0.40* + 0.25 .y < 125 Condition on fat I
(0.0) L
100 200 300
(312.5.0)
FIGURE 5 Corner points: (0, 0), (312.5, 0), (250, 100), (0, 300)
Since x and y represent numbers of pounds, we have x > 0, and y > 0. The
solution region is given in Figure 5>
EXAMPLE 6 A doctor has prescribed a diet in which the total number of calories must not
exceed 1200. She insists that twice the number of protein calories added to
the number of carbohydrate calories must equal or exceed 1600. Write these
conditions as a system of inequalities and graph the possible solutions.
SOLUTION Let x represent the number of protein calories and y represent the
number of carbohydrate calories. The conditions are
x + y < 1200
2x +y> 1600
> ,v
> >
The solution region is shaded in Figure 6>
FIGURE 6
Practice Problems > 1. Find the corner points for the system
x + y < 3
3.v - 2y < 4
x >
2. Find the solution region of
x + y < 2
x >
y >
and locate all corner points.
Exercise Set 3.1 Graph the solution regions, find the 11.
corner points, and classify the re-
gions as bounded or unbounded.
1. x
y =: u
2. x < 3
y<2
3. x >0
y>3
4. a:< 2
y >
5. y > x
jc >
6. y
y
7. y > x
a: >
y<3
8. y
y
y
9. y < at
x >
y<3
10. y < a:
y >
y<3
19. y >x - \ 21). x + y < 10

y < -x + \ Ax - ys 8
y< 1 6jt - 2y ^ 12
x > a: >
y > y>0
y<3
Applications
(Business and 21. Manufacturing. A manufactur- cal model, graph the solution re-
Economics) er makes two products, valves gion, and find the corner points
and reducers. A valve requires 1 for the investor.
hour on machine A and 2 hours
24. Assignment. The USX Coal
on machine B, and a reducer re-
Company operates two mines
quires 2 hours on A and 2 hours
with different production capac-
on B. Let x be the number of
ities. The following is the output
valves produced in a day and y
in tons of coal per day from each
the number of reducers pro-
mine.
duced in each machine
a day. If
operates up to 8 hours a day,
write the mathematical model
Quality Mine A Mine B
for this manufacturer, graph the
High-grade
solution region, and find the cor-
Medium-grade
ner points.
Low-grade
22. Manufacturing. Rework Exer-
cise 21 if the manufacturer de- For a week, the company needs
cides to operate each of the ma- at least 22 tons of high-grade, 15
chines up to 16 hours a day. tons of medium-grade, and 32
23. Investment. An investor has up tons of low-grade coal. How
to $30,000 to invest in AA bonds many days should the company
orB bonds. The AA bonds yield operate each mine in order to
8%, and the B bonds yield 12%. meet requirements? Shade the
If the investor wishes to invest solution region and find the cor-
at least 2 times as much in AA as ner points.

in B bonds, write the mathemati-
and Life Sciences) 25. Education. A classroom has boys and girls in a class and find
space for up to 50 students. Let the corner points. (Actually the
.v represent the number of girls solution region will consist of
and y represent the number of only the points with whole-num-
boys. Graph the region that rep- ber coordinates in the solution
resents the possible number of region.)
1S0 Chapter 3 Introduction to Linear Programming
26. Education. There are 210 stu- 28. Agriculture. George has a 50-
dents in the school referred to in acre farm on which he plans to
Exercise 25. The principal re- plant two crops, wheat and
quests that the number y of boys corn. Wheat requires 2 days of
and the number x of girls in the labor per acre, and corn requires
classroom satisfy the additional 3 days of labor per acre. The
requirement Ix + 3y < 210. other costs amount to $40 per
Graph the set of possibilities acre for wheat and $30 per acre
that would satisfy this condition for corn. George has evaluated
and the space condition of Exer- his assets and found that he has
cise 25. Find the corner points of up to 150 days of labor available
this set. and up to $1800 capital. Let x
represent the number of acres of
27. Education. The teacher in the
wheat planted and y the number
classroom of Exercise 26 insists
of acres of corn. Graph the solu-
that the number of boys in the
tion region and find the corner
classroom be at least 7. Use this
points.
additional restriction to graph
the set of possibilities and find
the corner points.
Answers to Practice > 1. The corner points are (0, 0), (0, 3), (2, 1), and (1$, 0).
Problems 2. The corner points are (0, 0), (2, 0), and (0, 2).
a- + >> = 2
3.2 Finding an Optimal Solution
Overview In the preceding section, we learned to locate a solution region. In this section we want to find
the largest and the smallest value of a linear function over the solution region. This study
presents the mathematical background for the graphical solution of linear programming prob-
lems in the next section. In this section we will introduce the following terms associated with
linear programming:
Objective function
Constraints
Finding an Optimal Solution 151
Feasible region
Optimal solution
We introduce the idea of optimal value by considering a simple selec-

tion process in the taking of a test.
EXAMPLE 7 On a test a multiple-choice question is worth 10 points and a true-false

question 4 points. It takes, on the average, 5 minutes to work a multiple-
choice problem and minutes to work a true-false problem. The test is 18

3
minutes in more than 4 questions. If you

length and you are not to answer
feel certain you know the answer to all problems, how many of each type
should you answer to get the best total score?
SOLUTION Let x be the number of multiple-choice questions that you answer

and v the number of true-false questions. Since you cannot answer more
than 4 questions,
x + y < 4
It takes 5.v minutes to work x multiple-choice problems and 3y minutes to

work y true-false. You are limited to 18 minutes, so
5jc + 3y < 18
In addition, since you can work only a positive number of problems (you
cannot work -3 problems),
x >
y >
In linear programming such inequalities as these four are called constraints.
The solution region for this set of constraints is shown as the shaded region
in Figure 7.
FIGURE 7
Note that the corner points are (0, 0), (^ 0), (3, 1), and (0, 4). (3, 1) is the
point of intersection of * +y = A and 5* + 3y = 18. Note also for this special
problem that only integral values have meaning, so the solution region con-
sists of a finite set of points: {(0, 0), (0, 1), (0, 2), (0, 3), (0, 4), (1, 0), (1, 1),
and (3, 1)}.
(1, 2), (1, 3), (2,0), (2, 1), (2, 2), (3, 0),
Remember problem is worth 10 points and a
that a multiple-choice
true-false question 4 points. If we answer x multiple-choice questions, we
have 10* points. Likewise, y true-false questions produce Ay points. The
total number of points we can attain is
P = 10* + Ay
We want to select x and v so that P is a maximum on our region of solutions.
One way to work this problem is to evaluate P at each of our 14 points in the
region of solutions.
(0,0) p =
(0,2)
(0,4)
(1, 1)
(1,3)
(2, 1)
(3,0)
5y, subject to the constraints, let us assign values to C and plot the lines
obtained. When C is assigned a particular value and the graph drawn, any
point (x, y) in the set of feasible solutions that lies on this line would produce
this same cost.
FIGURE 8
Thus, this line would be called a constant-cost line. As C takes on other

values, we obtain a family of constant-cost lines (see Figure 9) that are
parallel to each other, since they have the same slope. This cost function, in
slope intercept form, is
y = -2x + CIS
which has a slope of -2 and a y intercept of CIS. Note that for this cost
function, as C increases the y intercept increases, and the line moves up
from the origin. Also note that the last line to contain any values of the
feasible solution, C 5 is the line through the corner point (6, 6). Thus, C
,
assumes a maximum value at this point.
C= 10
6 + 5
6
= $90
In a like manner, C assumes a minimum value at (2, 3).
C= 10
2 + 5
3
= $35
c2 c, c4 c 5 c 6
FIGURE 9 *
Finding a maximum or minimum value of the objective function over a

feasible region is often called optimizing the function. The maximum or
minimum value of the function is the optimal solution.
Fundamental Theorem of Consider a linear function

Linear Programming
P = ax + by + c
over a feasible region defined by linear constraints. If this function has an optimal solu-
tion, it will occur at one (or more) of the corner points.
EXAMPLE 9 Find the maximum and minimum values of the function
P = Ax + 3y
subject to the constraints
x >
y >
5x + 3y < 30
2x + 7>y < 21
SOLUTION The constraint inequalities are shown in Figure 10, and the feasi-
ble region of solutions is shaded. The corner points are (0, 0), (0, 7), (3, 5),
and (6, 0). In 4x + 3v ory = (-4/3)* +

addition, the objective function, P=
/V3, with slope -4/3, is drawn for several values of P. Note that the maxi-
mum value of P seems to occur at (3, 5) and the minimum value at (0, 0). This
we verify by evaluating P at each corner point.
(0, 0):
SOLUTION From a graph of the constraint inequalities, Figure 11 shows that

the feasible region is not bounded. The only two corner points are (0, 0) and
(2, 0).
x + v = 2.v + v = -4
FIGURE 11
At the corner points
(0, 0): /= 2(0) + 3(0) = |

Minimum |
(2, 0): /= 2(2) + 3(0) = 4
However, let /take on values 6, 12, and 18, and draw the corresponding
lines.
Li
subject to
3x + 2y < 15
x + 3y < 12
x >
y >
2. Find the maximum and minimum values of
P = 3x + Ay + 6
under the constraints
2x + 3y < 12
-x + 2y < 1
a: >
y >
Exercise Set 3.2 Draw the region offeasible solutions 3. Maximize

and find the indicated optimal values
F =25x + 35y
and the corner points at which they
occur. If no solution exists, explain subject to
why. <
x + y 7
1. Maximize 2jc + y < 9
3jc + y < 12
P = 25x + 35y jc >
subject to y >
2x + 3y 15 Maximize and minimize
3x + y 12
P = 6x + 2y + 3
subject to
Maximize x + y >
2* + 3y < 6
P = 35x + 25y
x >
subject to y >
2jc +y = 7
3x + y < 8
x >
y >
5. Maximize and minimize subject to
F = I2x - 2y + 20 x + y > 2
2* + 3y < 15
subject to
3.v + y < 12
3x + 4y < 37 a: >
y> 2 y >
y < 4
10. Maximize and minimize
*< 3
x > P = 35jc + 25y
subject to
6. Maximize and minimize x + y > 2
F= 16a: - 4y + 20 2a: + y < 7

3a: + y < 8
subject to a: >
2x + 3y < 23 y >
y> 1
y^ 3
a: < 4 P = 25x + 35y
a:> 1
subject to
2 < a: + y < 7
7. Maximize and minimize 2a: + y < 9
3a: + y < 12
C= Ix + 2y + 4
>
a:
subject to y >
Ax + 3y > 24 12. Maximize and minimize
3* + 4y > 8
- 35y +
/> = 25jc 100
x >
y> 2 subject to
x + 3y < 21
8. Maximize and minimize 2a: + 3y < 24

2a: + y < 16
P = 2x + 3y + 2 a: >
subject to y >
2x + y > 10 13. Maximize and minimize

x + 2y < 8 P = 50x + 20y
jc > 2
y > subject to
2a- + 3y > 6
x + 3y < 21
2a: + 3y < 24
/> = 25a: + 35y 2a- + y < 16
a:, y >
Graphical Solutions to Linear Programming Problems 159
14. Exam. Lomax Fabricating Co. where x = the units of the first
makes only two products, with product, and y = the units of the
the following two production second product. If the profit
constraints representing two function is Z = $Sx + $4y. the
machines and their maximum maximum possible profit is
availability: (a) $40

(b) $42
2.x + 3y < 18
(c) $36
2x + y < 10
(d) $48
(e) Some profit other than those
given
Answers to Practice I 1. The maximum value of P is 75 at (3, 3).

Problems 2. The maximum value of P is 24, and the minimum value is 6.
5.5 Graphical Solutions to Linear Programming Problems
Overview Many practical problems in business, economics, life sciences, and the social sciences involve
relationships among variables that can be expressed linearly as inequalities. When this occurs,
we can find a region of feasible solutions and can thus find a maximum or minimum of some
objective function. In this section we examine a large number of application problems which are
called linear programming problems.
Let us now summarize some of the concepts introduced in the previous

section. A linear programming problem is a problem for which we are to find
a maximum or minimum value of a linear function in two variables P = ax +
by (in this section) or, in general, an n-variable linear function such as P =
C[X\ + c2 x2 + + cxn in the last part of the chapter. The linear function
we are to optimize is the objective function, and the variables (x and y in two
variables and x\ x 2 , , . .xn in n variables) are subject to constraints in the
. ,
form of linear inequalities. In addition, each of the variables must be non-

negative, Xi s 0, where (=1,2,. n. The region of solutions satisfying
. . ,
both the constraint inequalities and the nonnegative requirements is called

the feasible region of the problem. When the feasible region is bounded, the
objective function has both a maximum value and a minimum value in the
feasible region. When the feasible region is unbounded, maximum and mini-
mum values of the objective function may not exist.
The following steps should be helpful as we formulate a mathematical
model to represent a linear programming problem.
Solutions of Linear 1. Introduce the problem variables. (For difficult problems you may want to use the
Programming Problems problem-solving technique of guessing values before assigning variables.)
2. Express the objective function (that which is to be optimized) in terms of the variables
in the form
P = ax + by
3. Write the problem constraints using linear inequalities (and/or equations). For maxi-
mum-type problems, the inequalities are usually of the form
ex + dy s e
For minimum-type problems, the inequalities are usually of the form
ex + dyê
4. The variables are to be nonnegative
x>0, y>0
5. Find the graph, or the feasible solution region, of the system of constraints.
6. Locate the corner points by solving pairs of adjoining constraints changed to equa-
tions.
7. Evaluate the objective function at each corner point to determine the maximum or
minimum value.
We now follow these steps on both a maximum-type and a minimum-

type linear programming problem.
EXAMPLE 11 The parts division of the Central Manufacturing Corporation shapes all axles
tomeet specifications on two machines: A and B. A small-car axle requires 1
hour on machine A and 1.5 hours on machine B. A large-car axle requires 2
hours on A and 1 hour on B. If each machine operates 24 hours a day and the
manufacturer makes a $15 profit on large-car axles and a $10 an axle profit on
small-car axles, find the number of each of the axles that should be produced
for maximum profit.
SOLUTION Let's guess that we make 8 small-car axles (at a profit of $10
8)
and 6 large-car axles (at a profit of $15 6) for a total profit of
P = 10
8 + 15
6
= $170 i
See Table 1
Now replace our guess of 8 small-car axles with x small-care axles and our
guess of 6 large-car axles with y large-car axles. The profit-objective function
becomes
P = $10* + $15y
TABLE 1
Description Small Axles Large Axles
Number of each
(guess)
Profit from 10 15 6 = $170
guess
Number of each (variable)
Profit 10.Y I5y Limitation
on number
of hours
Hours needed 1 x hours 2y hours s 24
machine A
Hours needed 1.5.x hours 1 v hours < 24
machine B
Each small-car axle is on machine A for 1 hour, so x small-car axles

would utilize machine A for.v hours. Each large-car axle uses machine A for
2 hours, or y large-car axles use machine A for 2 y hours.
x + 2y < 24 Machine A operates up

to 24 hours each day.
Likewise,
1.5* + ly < 24 Each small axle needs

1.5 hours and each
large axle needs 1 hour
on machine B.
Can't have a negative

number of axles.
The feasible region is plotted in Figure 12, and the corner points are: (0, 0),
(0, 12), (12, 6), (16, 0).
At the corner points, the values of the objective function are
(0,0) P = 10(0) + 15(0) =
(0, 12) P = 10(0) + 15(12) = $180
(16,0) P = 10(16) + 15(0) = $160
(12, 6) P = 10(12) + 15(6) = $210
Thus, the maximum profit in one day on the two machines is $210 obtained
by producing 12 small axles and 6 large axles. Note that the minimum value
is $0, which is what one would guess>
FIGURE 12
EXAMPLE 12 A hospital dietician is experimenting to get a prescribed amount of calcium

and vitamin A and at the same time minimize cholesterol-producing units.
Each ounce of food A produces 20 units of calcium, 10 units of vitamin A,
and 6 cholesterol-producing units. Each ounce of food B produces 10 units
of calcium, 20 units of vitamin A, and 4 cholesterol-producing units. If the
minimum daily requirements are 300 units of calcium and 180 units of vita-
min A, how many ounces of each food are needed to meet daily require-
ments and at the same time minimize cholesterol-producing units? What is
the minimum number of cholesterol-producing units?
SOLUTION Let's guess 10 ounces of food A and 8 ounces of food B. The 10

ounces of food A would produce 6 10 cholesterol-producing units, and the 8

ounces of food B, 4 8 cholesterol-producing units. Thus, the number of

cholesterol-producing units would be 6 10 + 4 8 = 92. See Table 2.

TABLE 2
Description
The 10 ounces would give 20(10) units of calcium and the 8 ounces would
give 10(8) units of calcium for a total of
20(10) + 10(8) = 280 units of calcium
and
10(10) + 20(8) = 260 units of vitamin A

Now let's replace our guess of 10 ounces of foodby x ounces of food A A
and 8 ounces of food B by y ounces of food B. Then replace the 280 by a
lower bound, 300 units of calcium, and the 260 by a lower bound, 180 units of
vitamin A. That is,
20.v + lOy > 300

lO.v + 20y > 180
x >
y >
and the objective function becomes
C = 6x + 4y \C is cholesterol-
|
prod ucing u nits.
The feasible region of solutions and the corner points are found in Figure 13.
(0. 30)
30
FIGURE 13
(0, 30) C= 6(0) + 4(30) = 120
(14/2) C = 6(14) + 4(2) = 92 I Minimum

(18, 0) C = 6(18) + 4(0) = 108
The minimum number of cholesterol-producing units that meets the require-

ments for calcium and vitamin A is 92. Why is it reasonable that there is no
maximum number of cholesterol-producing units?^
Practice Problem [> 1. A sociologist wishes to maximize the time he can spend on his favorite
research project. He plans 14 morning sessions and 12 afternoon sessions
for a week. The sociologist does not want more than 3 morning sessions
and 2 afternoon sessions during 8 hours, nor more than 2 morning ses-
sions and 4 afternoon sessions during 8 hours. How long should each
session be in order for him to maximize the time spent on his project for
the week?
Answer to practice problem is found on page 169.
Exercise Set 5.3
Applications
(Business and Manufacturing. A manufactur- (b) Translate to a linear pro-
Economics) er makes two products, type I gramming problem by con-
and type II. Type I requires 4 structing the
hours on machine A and 3 hours (i) objective function and
on machine B; type II requires 2 the
hours on A and 3 hours on B. If (ii) system of constraints.
the machines operate up to 12 (c) Graph the system of con-
hours a day and the manufac- straints, shade the feasible
turer makes a profit of $14 for region, and find all corner
each unit of type I and $13 for points of this region.

each unit of type II, find the (d) Complete the solution of the
number of units of each product problem.
that should be produced for
maximum profit,
(a) First understand the prob-

lem by completing the fol-
lowing table.
Description Type I Type II
Number of each
(guess)
Profit from
(guess)
Number of each
(variable)
Profit from
(variable) Maximum number of
hours per day
Hours needed,
machine A
Hours needed,
machine B
2. Manufacturing. Redo Exercise $2000 per day to operate refinery

1 for profits of $80 per unit for II, how many days should each,
type I and $200 per unit for type be operated to satisfy the re-
II. quirements at minimum cost?
(a) First understand the prob-
3. Minimum Cost. An oil company lem by completing the fol-

requires 800, 1400, and 500 bar- lowing table.
rels of low-, medium-, and high- (b) Translate to a linear pro-
grade oil, respectively. Refinery gramming problem by con-
Iproduces 200, 300, and 100 bar- structing the
rels per day of low- medium-, (i) objective function and
and high-grade oil, respectively; the
refinery II produces 100, 200, (ii) system of constraints.
and 100 barrels per day of low-, (c) Find the feasible region and
medium-, and high-grade oil, re- all corners of this region.
spectively. If it costs $3500 per (d) Complete the solution of this
day to operate refinery I and problem.
Description Refinery 1 Refinery 2
Number of days
(guess)
Cost of operation
(from guess)
Number of days
(variable)
Cost of operation = C
(variable) Minimum
requirements
Low-grade oil
Medium-grade oil
High-grade oil
4. Assignment. The sanitation de- 6. Transportation. Redo Exercise

partment of the City of Clanton 5 with shipping costs of:
has 100 garbage trucks and 250 A to I, $8 per ton
employees. A full-strength col- B to I, $6 per ton
lection team consists of 1 truck
A to II, $5 per ton
and 3 employees. A full-strength
B toll, $10 per ton
team collects 12 tons of garbage
per day, and a partial-strength 7. Investment. An investor has up
team with 2 employees collects to $50,000 to invest in AA bonds
only 6 tons. The city manager or B bonds. The AA bonds yield
wishes to collect the maximum 10%, and the B bonds yield 15%.
amount of garbage each day, but If the investorwishes to invest
the numbers of operating trucks at least 3 times as much in AA
and available employees vary. bonds as in B bonds, find the
How many full-strength teams amount she should invest in
and how many partial-strength each type of bond to maximize
teams should be formed in order income.
to maximize the amount of gar-
8. Mixture. A shop sells a mixture
bage collected on a day when all
of cashews and pecans for $3 a
trucks are operating and a maxi-
pound. $1000 has been allocated
mum of 240 employees is avail- for buyingcashews at $2.00 per
able for assignment? Write the
pound and pecans at $1.50 per
mathematical model and graph
pound. The mixture must con-
the solution for the City of Clan-
tain at least twice as many pe-
ton's assignment problem; then
cans as cashews (by weight), but
find the corner points and the
no more than 3 times as many
answer.
(by weight). How many pounds
5. Transportation. A company has of each type should be ordered
two warehouses, A and B, and to maximize profit?
two stores, I and II. Warehouse
9. Manufacturing. A company man-
A contains 100 tons of a prod-
ufactures two types of electric
uct, and warehouse B contains
150 tons. Store I needs 50 tons
saws, M (medium size) and S
(small). It takes 2 hours to as-
of the product, and store II
semble each small saw and 3
needs 75 tons. The shipping
hours to assemble each medium-
costs are:
sized saw. The company can
A to I, $5 per ton work up to 800 hours per week.
B to I, $6 per ton However, the company can
pack and ship only 300 saws per
A to II, $8 per ton
week. If the company sells the
B to II, $10 per ton
small saw for $30 and the me-
Find shipping instructions that dium saw for $40, how many of
will satisfy the stores' needs at each type should be produced in
minimum shipping cost. order to maximize revenue?

10. Assembly A manufactur-

Line. If product A can be sold for a
ing company produces two profit of $8 each and product B
products, A and B. Both prod- for a profit of $6 each, how many
ucts require time on three as- of each would produce a maxi-
semblies. The following table mum profit? Remember
(Hint:
gives the time required in each to convert the maximum time to
assembly as well as the total minutes.)
time available in an assembly.
Product A Product B Maximum time
Assembly (minutes) (minutes) available (hours)
and Life Sciences) 11. Agriculture. George has found are gathered, at least 304 hours
that he can make $250 per acre are required for processing.
profit for each acre of wheat he Each instructor can process data
plants on his 50-acre farm (see for 8 hours and each graduate
Exercise 28 in Set 3.1). He can assistant for 4 hours. If instruc-
make $225 per acre profit for tors cost $10 per hour and gradu-
each acre of corn he plants. ate assistants cost $7 per hour,
Write his total profit as a linear find thenumber of each the uni-
function of the acres planted, versity should use for minimum
and use the corner points found cost. Assume that the university
in Exercise 28 to help George uses the same instructors and
decide how many acres of each graduate assistants to gather and
crop he should plant in order to process the data.
maximize his profit.
14. Education. Redo Exercise 13
12. Agriculture. Redo Exercise 11 for an instructor cost of $15 per

with a profit of $200 per acre for hour, where the graduate assis-
wheat and $300 per acre for tant's cost remains at $7 per
corn. hour.
13. Education. A university plans 15. Nutrition. A diet company is
to use instructors and graduate planning to market a diet prod-

assistants in a research project. uct composed of two foods.
The university needs at least 400 Food I contains 30 calories per
hours of labor spent in gather- unit, and each unit has 20 grams
ing data, with each instructor of protein, 20 grams of carbohy-
spending 8 hours and each grad- drate, and i grams of fat. Food II
uate student spending 20 hours contains 40 calories per unit,
gathering data. After the data and each unit contains 10 grams
of protein, 40 grams of carbohy- each must be purchased to mini-

drate, and 5 gram of fat. How mize cost and satisfy minimum
many units of each food should requirements?
the product contain if the com-
pany wishes minimize the
to 18. Nutrition (Plants). A farmer
number of calories, yet have the uses two types of fertilizer. A
mixture satisfy the requirements bag of 13-13-13 contains 7
of 60 grams of protein, 4 grams pounds of nitrogen, 7 pounds of
of fat, and 100 grams of carbohy- phosphoric acid, and 7 pounds
drate? of potash. 6-8-8 contains 4
pounds of nitrogen, 3 pounds of
16. Nutrition. A person on a low- phosphoric acid, and 4 pounds
carbohydrate diet plans to eat of potash. The minimum re-
two foods. A unit of food 1 con- quirements for a field are 2000
tains 5 grams of carbohydrate pounds of nitrogen, 600 pounds
and 100 calories, of which 10 of phosphoric acid, and 200
calories are protein. A unit of pounds of potash. If a bag of 13-
food II contains 6 grams of car- 13-13 costs $6 and a bag of 6-8-8
bohydrate and 100 calories, of costs $4, minimize the cost of
which 30 calories are protein. spreading fertilizer to satisfy
He wishes to minimize the num- minimum requirements.
ber of grams of carbohydrate
while eating at least 400 calories, 19. Exam. Lomax Company manu-
at least 60 of which are provided factures and sells two products,
by protein. Find the number of A and B. Time requirements and
units that he should eat of each selling prices for each product
type of food.
17. Nutrition (Animals). A calf feed

mixture is to be obtained from
peanut hulls and soybean meal. Products
The mixture must contain at Hours of labor for .0 1.5
least 100 pounds of protein, 40 each item

pounds of fat, and 10 pounds of Hours on a machine 0.5 2.0
minerals. Each 100 pounds of for each item

Selling price of each $27.50 $75.00
peanut hulls contains 15 pounds
item
of protein, 5 pounds of fat, and 1
pound of minerals. Each 100

pounds of soybean meal con-
tains 50 pounds of protein, 10 The company can work a total of
pounds of fat, and 4 pounds of 600,000 hours of labor and can
minerals. If each 100-pound bag use up to 200,000 hours on ma-
of peanut hulls costs $8 and each chines. The following questions
100-pound bag of soybean meal relate to the use of linear pro-
costs $12, how many bags of gramming to maximize gross
sales, where x is the number of (C) 8* + 12 v 200.000

----
product A
produced, and v is the (d) I* + l.5v < 4,800,000
number of product B produced. (e) I* \ 1.5? < 600,000
The objective function is
21. Exam. For 19,Exercise
the
(a) [0.50* + 32.00v
constraint function for machine
(b) 8.50* + 24.00v
capacity is
(c) 27.50* 4 75.00v
(a) 6* + 24? < 200,000
(d) 19.00* + 51.00?
(b) 0.5* + 2? < 200,000
ie) [7.00* + 43.00?
(c) (0.5 + 1)* + (2 + 1.5)? <
20. Exam. For the preceding prob- 200,000
lem, the constraint function for (d) 0.5* + 2? < 600,000
labor is Lv _L5?
(e) < 800,000
(a) I* + 1.5? < 200,000
(b) 8* + I2y < 600,000
Answer to Practice 1. Let * be the length in hours of morning sessions and ? the length in hours
Problem of the afternoon sessions.
3* + 2? s 8
2* + 4? < 8
* > 0, ? >
T = 14* + 12?
The morning sessions should be 2 hours long, and the afternoon sessions
1 hour long, to maximize the time spent on the project.
3.4 A Graphical Introduction to the Simplex Method
Overview In the preceding sections we considered the geometric method of solving linear programming
problems. Practically speaking, this procedure is useful only for problems involving two, or at
most three, variables. However, our study of geometric procedures provides an excellent back-
ground for the study of the algebraic procedures that are needed for linear programming
problems involving more than three variables. In this section, we introduce an algebraic ap-
proach called the simplex method. This procedure makes use of our knowledge of row opera-
tions on matrices. Such a procedure is readily extended to any number of variables, and with
computers linear programming problems involving hundreds, and even thousands, of variables
can be considered. In this section, we will consider
Slack variables
Basic feasible solutions
Pivot operations
The simplex method
The simplex method (developed by George Dantzig in 1947) is a proce-

dure for solving algebraically maximum-type linear programming problems.
Let us return now to the manufacture of axles discussed in Example 1 1
Maximize
P = lOx + 15y Objective function
subject to
x + ly < 24 Problem constraints (1)

1.5* + ly < 24
x > Nonnegative constraints
y >
where x and y are the number of small-car axles and the number of large-car
axles, respectively, produced each day. Notice that all the problem con-
straints involve "less than or equal to" inequalities with positive constants
on the right. In the next three sections we consider only maximization prob-
lems that satisfy this condition. Such a problem is sometimes called the
standard maximization problem.
Standard Maximization A linear programming problem is said to be a standard maximization problem if it can be
Problem written in the following form:
Maximize
P = CfXt + c2 x2 + + cn x n Objective function

A Graphical Introduction to the Simplex Method 171
subject to
Although we omit the details, it should be fairly obvious that the sys-
tem of inequalities (1) and the system of equations (2) are equivalent in the
sense that if x = a and y = b is any solution of (1), then there are values of
the slack variables so that x = a, y = b, r = c, and s = d is a solution of (2).
That is, x = 2, y = 4 satisfies (1) and x = 2, y = 4, r = 14, and s = 17 satisfies
(2). Conversely, if x = a, y = b, r = c, and s = d is a. solution of (2) then
x = a, y = b is a solution of (1).
Next we write the objective function
P = lOx + 15y
in a form with all the variables on the left and on the right.
IOjc - 15y + P =
We then have the following system of equations:
x + 2y + r = 24
l.5x +y + s = 24
-\0x - I5y + P =
Each solution of this system is a set of numbers
x, y, r, s, P
Now unknowns, an infinite number of
recall that for three equations in five
solutions may by assigning values to two of the variables and solving
exist;
for the other three, you can obtain a solution (provided that a solution
exists). When we set two of the variables equal to and solve for the other
three, the answer is called a basic solution. In general, with m equations and
n unknowns where n > m, we set n - m variables equal to zero and solve for
the remaining in variables to get a basic solution.
Definition: Basic Feasible When a linear system consisting of m equations and n unknowns such that n >m is
Solution associated with a linear programming problem, any basic solution of the system that has
no negative values is called a basic feasible solution. The n - m variables set equal to
in obtaining a basic solution are called nonbasic variables; the m variables for which we
solve are called basic variables.
EXAMPLE 14 Maximize
P = Ax + 3y
subject to
5x + 3y < 30
2x + 3y < 21
x, y s=
SOLUTION Using slack variables, the corresponding problem as a system of

equations is:
Maximize
P = Ax + 3y
subject to
5x + 3v + r = 30
2x + 3y + s = 21
x, y, r, s S:
In the system in Example 14, suppose we set two variables equal to

and solve for the other two variables. There are six basic solutions:
x =
points. Since the objective function does not contain r and s, evaluating the
objective function at the basic feasible solutions is equivalent to evaluating
the objective function at corner points. Thus, we can obtain the maximum
value of the objective function by evaluating at the basic feasible solutions:
.v = p = 4 + 3-0 = (3)
P = 4 + 3 7 = 21

(5)
P = 4
6 + 3 = 24

(6)
P = 4
3 + 3 5 = 27

(8)
Thus, the maximum value of P is 27. This suggests, without proof, the
following fundamental theorem.
We have already noted that a corner point corresponds in a natural way

to a basic feasible solution.We now show how to move from one corner
point (in the tableau) to anew corner point (represented by another tableau).
By this procedure, a search can be made for the corner point that will
optimize the objective function. To accomplish this goal, we use elimination
procedures of matrix theory. Our goal is to obtain the identity matrix as the
coefficient matrix of different variables. This will give anew corner point of
the basic feasible solutions. For example, we perform row operations on
Tableau 9 to get a new tableau where the identity matrix is the coefficient
matrix of the variables x, s, and P instead of r, s, and P.
X
5
2
(9)
make the pivot element 1 and to make other entries in the column containing
the pivot element (called the pivot column) 0.
Definition: Pivot Pivot operations are row operations (other than interchanges) performed on a matrix to
Operations make the pivot element 1 and to make all other entries in the pivot column 0.
EXAMPLE 15 Perform a pivot operation on the following simplex tableau with pivot ele-
ment 3 (circled) to obtain a second matrix. Write the system of equations
represented by the second matrix.
SOLUTION
Need a g/?l - R\
Need a
L-14 0J
subject to
3x + 2y
5x + y
x, y
Write a corresponding problem using equations.

2. Form a simplex tableau for the information given in Practice Problem 1.
3. Perform a pivot operation on the simplex tableau in Practice Problem 2 to
get a second tableau if the pivot element is 3 in the first row and first
column.
4. Using the simplex tableau
X
2
4
6. Maximize 11. Maximize

P = 35x + 25y P = 35x - 25y
subject to subject to
2x + y < 7 x + y < 7
3x + y < 8 2* + y < 9
3jc + y < 12
12. Maximize
7. Maximize P = 25.x + 35y
P = 35jc - 25y subject to
subject to x + 3y < 21
2x + 3y < 15 2* + 3y < 24
3* + y < 12 2jc + y < 16
13. Maximize
P = 25x + 35y
8. Maximize
subject to
P = 25* - 35y
2x + 3y < 6
subject to + 3y < 21
jc
2x + 3y < 24
2jc + 3y < 15
2* + y < 16
3x + y < 12
14. Maximize
P= 25* + 35y
9. Maximize
subject to
P = 35x + 25y
<
x + y 2
subject to x + y < 7
2jc + y < 9
x + y < 7
3* + y < 12
2* + y < 9
3x + y < 12 15. Listed below are all the basic
solutions for the system
+ y + r =
2x 7
10. Maximize + y + s =
3x 8
-35* - 25y + P =
P = 25* 4- 35y
For each basic solution, identify
subject to
the nonbasic variables and the
x + y < 7 basic variables; then, classify
2* + y < 9 each basic solution as feasible or
3jc + y < 12 not feasible.
(a) x =
25. X
~1
26.
The Simplex Method of Maximization 181
Answers to Practice 1. Maximize

Problems
P = lx + 3y
subject to
3x + 2y + r = 4
5x + y + s = 8
x, y, r, s ^
V
3
5
Let's return now to a linear programming problem we have studied.
Maximize
P = 14* + 12y Objective function
subject to
3x + 2y < 8 Problem constraints

2x + 4y < 8
x, y > [
Nonnegative constraints
with corresponding tableau
X
3
2
Procedure for Finding the Choose as the pivot column the variable column that has a negative number in the
Pivot Element bottom row that is the largest number in absolute value. If the negative number in the
last row with the largest absolute value occurs in two or more variable columns,
choose any one of these columns as the pivot column and proceed to the next step.
Divide each positive element above the dashed line of the pivot column into the right
column entry in the same row of the positive element. If the pivot column has no
positive elements above the dashed line, then there is no solution.
Choose as the pivot row the row for which the quotient obtained in step 2 is smallest.
If two or more quotients are the same, choose any one of these rows as the pivot row.
The pivot element is the element in both the pivot row and the pivot column and is
denoted with a circle.
Note that the pivot element is always positive and is never on the bottom
row.
Our next step will be to perform appropriate matrix row operations
such that our pivot element will be and the remaining elements in the pivot
1
column will be 0.
v r s P
Need a 1
[NeedaOj-
rr -> 2
2
4
10
10 -2R + X
/?;-/?; (2)
-12 o""i OJ 14/?, + Ry-> R 3 (3)
The second tableau is
X
The third tableau is
P cannot be made larger by increasing any of the variables. This is seen by

writing the equation represented by the last row of the matrix:
4r + s + P = 40
The Simplex Method of Maximization 18S
bx + 5y < 960
lO.v + lly < 1760
SOLUTION
Step 1. Introduce slack variables and write the resulting equations.
6.v + 5y + r = 960
lftr + \\y + s = 1760
Step 2. Form the first tableau.
V
The new simplex tableau with the pivot element indicated is
X
2. Maximize 7. Maximize
P = 2x + 3y P = 35.v, + 25.v 2
x > x, >
y > x2 ^
X + y < 4 2.vi + jc 2 ^ 7
3x + 2y < 9 3.xj + a: 2
< 8
3. Maximize
8. Maximize
P = 5x + 4y
P = 35*, + 25x2
subject to
subject to
jc>
y > -v, >
.r + >> < 4 X2 3=
3.r + 2y < 9 2.v, + 3a: 2 s 15

3jc, + x2 ^ 12
4. Maximize
P = 4x + 3y
9. Maximize
subject to
P= jc + 8y + 9z
x >
y
> subject to
jc + y < 4
3.v + 2y < 9
jc>0
5. Maximize v>0
P = 25x + 35y
zÔ
5* - 2y - 3z <
subject to -3* + y + z ^
> -5x + 3y + 4z < 200
;t
y >
2.v + y < 7
2* + y < 8 10. Maximize
P = x - 3y + 2z
6. Maximize
subject to
P = 25.v + 35y
x >
subject to >
y
x >
y > .y + 6y + 3z < 6
2x + 3y ^ 15 .v + 2y + 4z < 4
3* + y < 12 jc - y + z < 3
11. Maximize 13.
P = 2x + 9y + 5z
subject to
* >
v s
z >0
3*+ 2y - 5z < 12
-x + 2y + 3z = 3
* + 3y - 2z < 2
12. Maximize
P = 7.3* + 5.2y
subject to
x >
y >
1.3* + 2.4y < 25
3.5* + 4.6? < 30
5000 gallons of high-sulfur must invest at least twice as

crudes. Available tonnages for much money in good-quality as
filling this order are 100 tons of it invests in high-risk bonds.
coal and 150 tons of shale. The
costs of the solids per ton are
20. Mixture Problem. A nut com-
pany has at most 64 pounds of
$20 for coal and $25 for shale.
pecans and at most 132 pounds
The processor receives $0.50
of peanuts that it wishes to mix.
per gallon for low-sulfur crude
and $0.30 for high-sulfur crude.
A package of mixture I contains
2 ounces of pecans and 11
17. Manufacturing. A factory man- ounces of peanuts, and a mix-
ufactures two items, A and B. ture II contains 8 ounces of pe-
Sixty cubic feet of gas and two cans and 12 ounces of peanuts
days of labor are required to fab- per package. If a profit of $0.64
ricate item A. Item B requires 40 per package obtained from
is
cubic feet of gas and 3 days of mixture and a profit of $0.70

I
labor. Because of a severe win- per package is obtained from

ter, only 1480 cubic feet of gas mixture II, how many packages
are available to the factory each of each mixture should be made
week. Sixty-six days of labor are to obtain the maximum profit?
available weekly. If item A pro-
duces a profit of $36 and item B 21. Advertising. Modern Electron-
ics has $40,000 available for ad-
yields a profit of $40, how many
of each item should be manufac-
vertising. The following table
gives the cost of an advertising
tured weekly to maximize
profit?
package and the number of peo-
ple in the advertising audience
18. Investments. An investment for each advertisement in two
club has at most $27,000 to in- different media (numbers in
vest bonds of two types,
in thousands).
good-quality and high-risk.
Good-quality bonds average a
Newspaper TV
yield of 7%, and high-risk yield
10%. The policy of the club re- Cost $ 4 $10
amount invested
quires that the Audience 20 60
in more than
high-risk be not
twice the amount invested in
If the maximum number of
good-quality. How should the
club invest its money to receive
newspaper advertisements is 8
the maximum return subject to
and the maximum number of tel-
its investment policy? evision advertisements is 2, how
many advertisements of each
19. Investments. Redo Exercise 18 type should be purchased to
assuming that a conservative maximize advertising exposure?
member of the club gets the club (Complete the solution of Exer-
policy changed so that the club cise 29, Set 3.4.)
and Life Sciences) 22. Agriculture. A farmer has avail- The old process releases 12
able 200 acres on which to plant grams of pollutant A and 30
either wheat or corn or both. grams of pollutant B into the air
The planting costs (seed, fertil- for each liter of chemical pro-
$150 per acre for
izer, etc.) are duced. The new process re-
corn and $60 per acre for wheat. leases 5 grams of pollutant A
The farmer has only $16,500 and 15 grams of pollutant B for
capital to spend on planting each liter produced. The com-
costs. If the anticipated income pany makes a profit of $0.40 for
is $420 per acre for corn and each liter produced by the old
$250 per acre for wheat, how process and only $0.18 for each
many acres of each should be liter produced by the new pro-
planted to maximize the in- cess. If the regulations do not al-

come? low more than 10,000 grams of
pollutant A and no more than
23. Pollution. Government regu- 27,000 grams of pollutant B each
lations for pollution control have day, how many liters of the
forced Langley Chemical to in- chemical should be produced by
stall a new process to help re- the old process and how many
duce pollution caused by pro- liters should be produced by the
duction of a certain chemical. new process to maximize profit?
Answers to Practice 1. The pivot element is 5.
Problems
X
Minimization Using the Dual Problem 191
).(> Minimization Using the Dual Problem
Overview In our use of the simplex procedure in the preceding sections, our constraints have satisfied
conditions that we classified as standard maximum linear programming problems. In this sec-
tion, we consider problems that do not satisfy these conditions. First, we want to consider
problems where we minimize the objective function. Associated with each maximum problem is
a minimum problem called its dual. Similarly, each minimum linear programming problem has a
corresponding maximum problem, also called its dual. To solve a minimum problem, we simply
use the simplex method to solve the dual maximum problem. In this section, we will study
Formation of the dual problem
Solution of the minimization problem
As we just mentioned, there is associated with each linear program-

ming problem a dual problem. For instance, the following are dual problems.
Maximize
we will be solving the minimum problem in terms of the maxi-

Since
mum problem using the simplex method for the standard maximum prob-
lem, we hereby outline how to change a minimum problem to a maximum
problem.
EXAMPLE 17 Minimize
C = 2y, + 5y 2
subject to
3y, + 2y 2 > 8
yi + 4y 2 s= 6
yi^Q
SOLUTION
Minimum Problem Maximum Problem
min C = 2y + 5y 2 t
max P = 8*1 + 6*2
3yi + 2y 2 &
3
3a:,
2*i
+ x2
+ 4*2 =s 5
=s 2
y
yi , y2 s *| , x2 >
Coefficient Coefficient
Matrix Matrix
One matrix is the transpose

of the other.
Thus, the tableau for the given minimum problem is the tableau for the dual
maximum problem.
(D
2
Perform the following row operations

Practice Problems > 1 . Write the simplex tableau for the standard maximization problem in order
to solve the following:
Minimize
C= 2y, + 5v 2
subject to
3y, + 2y 2 > 8
y, + 4y 2 > 6
2y, + },>4
y\ , y2 a o
A simplex tableau for a minimization problem is
*1
Minimization Using the Dual Problem 19S
5. Exercise 1
6. Exercise 2
7. Exercise 3
8. Exercise 4
Write the linear programming prob-

lems and duals that are represented
by the following simplex tableaus:
*l
10.
19. Minimize 20. Minimize
25;v, + 35 v: = C 25.v i
+ 35 y 2 = C
vi + y2 > 2 2y, + 3y 2 a 6
2.v, + 3y 2 ^ 15 2y, + 3y 2 > 15
3yi + y2 s 12 3.vi + y 2 2= 12
y, so y, >
y 2 >0 y2 >0
Applications
(Business and 21. Transportation. A company has tamin B, and 4 units of vitamin
Economics) two warehouses, A and B, and C. Product S costs 15 cents per
two stores, I and II. Warehouse gram, and each gram contains 10
A contains 40 tons of a product units of vitamin A, 6 units of vi-
and warehouse B contains 100 tamin B, and unit of vitamin C.
1
tons of the same product. Store I How many grams of each prod-
needs 50 tons of the product and uct should the company use in
store II needs 75 tons. The ship- the mixture in order to minimize

ping costs are: costs?
A to I, $5 per ton 23. Production. Acme Pharmaceu-

B to I, $6 per ton ticalproduces two strengths of a
newly developed drug. (Drug A
A to II, $8 per ton
is of greater strength than B.)
B to II, $10 per ton
Each costs $1000 per gallon to
Find the shipping instructions produce. It has been determined
that will satisfy each store's that at least 30 gallons of A and
need at a minimum shipping at least 20 gallons of B must be
cost. Solve graphically. produced within the next month.
Drug A requires one pound of a
22. Mixture Problem. The Acme
perishable raw material per gal-
Pharmaceutical Company wish-
lon, and drug B requires two
es to manufacture a vitamin mix-
pounds. The current inventory
ture so that each bottle contains
of this perishable raw material is
at least40 units of vitamin A, 22
80 pounds, and this much per-
units of vitamin B, and 22 units
ishable raw material must be
of vitamin C. The mixture will
used within the month. Find the
be made from products R and S.
number of gallons of each to
Product R costs 10 cents per
produce in order to minimize to-
gram, and each gram contains 5
tal cost.
units of vitamin A, 2 units of vi-
and Life Sciences) 24. Nutrition. George Ratshaw is be protein. Each food contains
on a special diet, which requires 100 calories per unit, but food I
a minimum number of grams of contains only 30 protein calories

protein and carbohydrates. He per unit, whereas food II con-
must have at least 40 grams of tains 50 protein calories per unit.
protein and 43 grams of carbo- Help Abdul decide how many
hydrates each meal. For lunch units ofeach food he should eat
he will eat food A. which costs to minimize the amount of car-
$2 per unit, and food B, which bohydrate.
costs $1.70 per unit. A unit of A 26. Medicine. Dr. Jones has de-
contains 8 grams of protein and
cided her patient needs at least
6 grams of carbohydrates. A unit
14 milligrams of drug I and at
of B contains 5 grams of protein
least 16 milligrams of drug II
and 7 grams of carbohydrates.
each day. These drugs are to be
What is the cheapest meal he
obtained by taking medicine I
can eat (with foods A and B) that
and medicine II. Both medicine
contains the minimum require-
I and medicine II contain the un-
ments?
desirable drug X. One gram of
25. Nutrition. Abdul is on a low- medicine I contains 3 milligrams
carbohydrate diet. He is plan- of drug I, milligram of drug II,
1
ning a meal composed of two and 3 milligrams of drug X. One

foods: food I with 7 grams of gram of medicine II contains 2
carbohydrate per unit and food milligrams of drug I, 2 milli-
II with 4 grams of carbohydrate grams of drug II, and 2 milli-
per unit. In order to keep him grams of drug X. How many
from becoming discouraged with grams of medicine I and medi-
his diet, Abdul's doctor has in- cine II should Dr. Jones pre-
sisted that he consume at least scribe each day if she wishes to
500 calories ateach meal, of minimize the amount of drug X?
which at least 210 calories must
Answers to 1. Maximize
Practice Problems
P = 8.v, + 6x 2 + 4.v 3
*l
2
5.7 Mixed-Constraint Linear Programming
Overview The standard form of a maximization problem involves constraints written as
aix-i + a2 x2 +
+ ax n s b bÔ
with objective function
P = Mi + k2 x 2 +
+ Mn
The standard form of a minimization problem involves constraints
ciyi +c 2 y2 + + c n y n ^ cf d>0
with objective function
C = /),/, + h 2 y2 +
+ hyn
In this section we will work problems that do not exactly fit either of these categories. That is, we
will consider mixed-constraint linear programming problems. First we will consider a mixed-
constraint problem geometrically. Then we will use the simplex procedure to solve the problem.
Consider the following example involving both < and s in the con-
straint equations.
EXAMPLE 18 Maximize
P = 4*i + x2
subject to
x\ + x2 ^ 8
-x + x2 s
{
4
x\ , x2 s
SOLUTION Geometrically the feasible solution and corner points are shown in
Figure 15.
The maximum value of the function occurs at (2, 6).
Maximum P = 4(2) + 6 = 14
Let's solve the preceding example using the simplex procedure. In

typical fashion we write the first equation as
jci + x2 + Si = 8
where s\ is our usual slack variable. We now introduce a second nonnegative

variable r 2 and subtract it from the left side of the second inequality to get
.V| + x 2 r2 = 4
Mixed-Constraint Linear Programming 199
FIGURE 15
This variable is not a slack variable because we add slack variables to get
equality. The variable r> is called a surplus variable because it is the amount
or surplus by which the left side of the equation is more than 4. Just like
slack variables, surplus variables are always nonnegative.

This linear programming problem can now be expressed as
x\ + x2 + s\ = 8
-.V, + x2 - r2 = 4
-4jc, - x2 + P =
X\, Xz, Si, Tj. S

We stop now to do some checking to see if this system satisfies our
requirements for a solution. First, let's look for a possible basic feasible
solution when x\ = and x 2 = 0. The solution is
X\ 0, x2 = 0, Si = 8, r2 = -4, and P =
This does not satisfy the requirement, r 2 > 0. Therefore, the simplex proce-
dure will not work on this set of equations.
To overcome this difficulty we introduce another nonnegative variable
a 3 such that
x\ + x 2 - r2 + ay = 4
The a 3 is often called an artificial variable because it has no real meaning
relative to the original set of constraints. It will have a value that will allow r2
to remain positive. We must also introduce this variable in the objective
function, which we write as
P = 4*i -I- x2 - Kaj K considered as large and >

We have now modified our system to the following:
X\ + X2 + S] = 8
-X\ + x2 - r2 + a3 =4
-4*i - x2 + Kai + P =
X\ , x2 , S] , r2 , 3 ^
Now setting x\ = 0, x2 = 0, and a 3 = 0, we see that r2 is still negative. To
solve this difficulty we eliminate K under a 3 in the last line of the following
tableau:
-1
As this example illustrates, if a 3 = in the optimal solution of the
modified problem, then deleting a 3 produces the same optimal solution in the
original problem. What happens if a 3 + 0? It can be shown that the original
problem has no optimal solution.
Sometimes one of the inequalities in a system of constraints is replaced
by an equation. Usually when this occurs, one can solve the system by using
the equation without a slack variable. If this does not yield a solution, one
can try introducing surplus and artificial variables.
EXAMPLE 19 Maximize
P = 2jc, + 3* 2
subject to
xi + x2 ^ 4
*i + 2*2 = 6
*i ,.r : >0
SOLUTION The system of equations can be written as
X] + x2 + si = 4
X\ + 2*2 r2 + a) = 6
-2*, - 3*2 + Kai + P =

This system can be represented by the following tableau:
X|
Using i in the first row, first column as the pivot element, we get
X, x2
1
Possible Steps for Mixed

Problem Constraints
SOLUTION Notice that the first inequality constraint is stated as less than or
equal to instead of greater than or equal to as required in order to set up an
initial matrix. To change the sense of the first inequality, multiply each term
by -1. The problem then becomes:
Minimize
7 = 2>-, + 5y 2 + 6)< 3 + 1
subject to
3yi - 2y 2 + 4y 3 > -
y, + 2y 2 + 2y 3 > 2
2ji + 3y 2 + )"3 S 5
The transpose of
the simplex tableau. Remember that we assumed that all of the elements of
the last column were nonnegative with the possible exception of the element
in the last row and last column. A special starting procedure is required for
problems in which some of the constants in the last column other than the
constant in the last row and last column are negative.
Exercise Set 5.7 Solve graphically the following 5. Minimize

mixed-constraint problems:
C = 3*i + 8*2
1. Maximize
subject to
P= 8*1 + 10jc 2
3*i + *2 < 9
subject to *l + 3*2 5: 6
+ 40
*, , *2 S
8*, 5*, =
4*, + 3*2 > 12 6. Minimize

*, , *2 s
C= 4*, + 3*2
2. Maximize
subject to
P = 10*, + 8*2
3*i + 4*2 ^ 24
subject to 5*i + 3*2 ^ 15
< *, *2 2=
8*, + 5*2 40 ,
6*i + 2*2 = 36 FmJ /toe initial simplex tableau in

*, , *2 s terms of surplus and artificial vari-
ables as if you were solving the fol-
3. Minimize
lowing problems. Don't forget to
P = 4*, + 10*2 eliminate K for the last row.
subject to 7. Maximize
8*, + 5*2 = 40 P = 8*, + 4*2
4*i + 3*2 s 12
subject to
*1 , *2 ^
*, + 2*2 ^ 8
4. Maximize
Xi+ * 2 > 2
P = 5*, + 4*2 *i * 2 a ,
subject to 8. Maximize
*, + 2*2 ^ 4 P = 10*i + 8*2
-*i + x2 > -6
subject to
*1 , *2 ^
*, + 3*2 ^ 12
*1 + *2 = 8
*1 , *2 2:
9. Maximize 13. Minimize
P = 5*, + 7*2 C = 2*, + 3*2
-X] + 2* 2 s 4 *, + *2 s 8
*, + *2 = 6 3*1 + 2*2 > 6
JCi , *2 S: *i ,x2 >0
10. Maximize 14. Minimize
P = 5*, + 3a: 2 C = 5*i + 7*2
*, - 2a: 2 ^ 4 -*, + X; > "2

jci -t- *2 ^ 6 *, + *2 < 4
*! , *2 2= *| , *2 ^
11. Maximize So/ue //ze following problems using
the simplex techniques involving
P = 4*, + 6*2
surplus and artificial variables:
subject to
15. Exercise 1
-*, + * 2 & -2
16. Exercise 2
*, + *2 > 4
*1 , *2 2: 17. Exercise 3
12. Maximize 18. Exercise 4
5
Z = lO*, + 12*2 19. Exercise 5
subject to 20. Exercise 6
-*, + 2*2 2: -4
*, + *2 > 6
*i , *2 2:
Applications
(Business and 21. Resource Allocation. USZ Min- pany has contracted to provide a
Economics) ing Company operates two local smelter a minimum of 14
mines, A and B. Each mine pro- tons of high-grade ore per week
duces high-grade ore and low- and a maximum of 6 tons of low-
grade ore. Mine A produces 2 grade ore per week. Determine
tons of high-grade ore and 0.5 the minimum cost to meet this
ton of low-grade ore per day. contract if it costs $4000 per day
Mine B produces ton of each 1 to operate mine A and $3000 a
grade of ore per day. The com- day to operate mine B.
and Life Sciences) 22. Nutrition (Plants). A fertilizer pounds of each. If there is a
company has available 10 tons profit of 20 cents on each 50-
of nitrate and 8 tons of phos- pound bag of A and a profit of 15
phate to produce two types of cents on each 50-pound bag of
fertilizer.A bag of fertilizer A B, how many of each would be
will have 20 pounds of nitrate mixed for maximum profit if all
and 10 pounds of phosphate, and of the phosphate must be used in
a bag of fertilizer B will have 10 this mixing endeavor?
Linear programming is used extensively in a variety of real world prob-

lems. The following examples have been reported in Interfaces.
Frontier Airlines used linear programming to manage individual aircraft

fueling on short-haul trips
Central Carolina Bank used linear programming to incorporate market
and regulatory constraints in managing loans and investments
A Sri Lanka coconut processor used linear programming to plan produc-
tion levels of desiccated coconut, copra, oil, and other products
The Chessie System saved over $2 million in just one year by using linear
2
programming to schedule box car repairs and acquisitions
Marketing
When we buy a new pair of tennis shoes, visit our favorite grocery store, or
drink our evening cola, it is difficult for us to realize how much of the cost of
the purchase represents the cost of advertising. The Oldsmobile Division of
General Motors Corporation spent $38 million on advertising in 1988. In
addition, a medium-sized dealership, Drennen Oldsmobile, spent more than
$480,000 on advertising. To get the most productive advertising for the
money spent, Drennen Oldsmobile applied linear programming techniques
to marketing or advertising. Drennen utilizes three avenues for advertising:
television, newspaper, and radio. Of course, a comprehensive study would
involve many classifications of these three; however, to simplify the problem
for this example, we consider that an advertisement will run during a given
week and no other advertisement will be used simultaneously. In addi-
that
tion, we assume that all television advertising is purchased for semi-
will
prime time ($800 for a 30-second advertisement), that newspaper advertising
:
Laplin, Lawrence. Quantitative Methods for Business Decisions, 4th ed., 1988, Harcourt
Brace Jovanovich. Inc.. Orlando, Florida, p. 207
is for a half page ($8000 for a week), and that radio advertising consists of
prime-time announcements at 5 radio stations ($6000 per week). With these

assumptions, advertising costs were
$16,000 per week for television

$8000 per week for newspaper
$6000 per week for radio
Advertising response is indicated by the number of prospects per week

(excluding unusual times such as Christmas week).
Medium Number of Prospects
Television 25-35 (average 30)

Newspaper 12-18 (average 16)
Radio 1 1-13 (average 12)
Using the average as a typical response, Drennen Oldsmobile computed the

number of weeks of advertising for television, newspaper, and radio that
would maximize the number of prospects per year; that is, Drennen com-
puted the maximum P for
P = 30x + 16y + I2z
where .v number of weeks of television, y the number of weeks of

is the
newspaper, and z the number of weeks of radio advertising, subject to the
constraints:
x + y + z ^ 52 Weeks in a year
$1 6,000 .v + $8000 v + $6000 z < $480,000 Annual budget
.v < 15 Not more than half of the

budget should be used for tele-
vision.
The simplex tableau for this problem was
X
1
Pivot element (175) Dual problem (191)

Pivot row (182) Solution of minimization problem
Mixed constraints (198) (191)
Pivot column (176)
Review Exercises Sketch the solution set of the follow-

ing systems of linear inequalities.
Locate the corner points. Is the solu-
tion set bounded?
14. A person on a low-carbohydrate 15. A chicken-feed manufacturer

diet plans to eat two foods. A mixes dried algae and crushed
unit of food I contains 5 grams of eggshells together and sells the
carbohydrate and 100 calories, mixture under the trade name
of which 10 calories are protein. "Start-Up." Each bag of Start-
A unit of food II contains 6 up must contain at least 52 units
grams of carbohydrate and 100 of vitamins and 52 units of min-
calories, of which 30 calories are erals. Each kilogram of dried al-
protein. The individual wishes gae costs $0.65 and contains 5

to minimize the number of units of vitamins and 2 units of
grams of carbohydrate while minerals. Each kilogram of
eating at least 400 calories, of crushed eggshells costs $0.28
which at least 60 are protein cal- and contains 1 unit of vitamins
ories. Find the number of units and 3 units of minerals. How
that the individual should eat of much of each ingredient should
each type of food. be in each bag in order to mini-
mize cost?
Chapter Test
1. For basic feasible solution of the

simplex tableau
2x < 8 - 3y
4.x + y < 6 X
2y < 8 - 6x
x, y >
graph the solution region, find
the corner points, and classify
the solution region as bounded
or unbounded.
2. Find the maximum and mini-

mum of
P = \0x + 20y
under the constraints of Exer-
cise 1.
3. Set up an initial tableau to maxi-

mize P = 10jt + 20y under the
constraints of Exericse 1.
4. Locate the pivot element in Ex-

ercise 3.
5. Determine the basic variables,

the nonbasic variables, and the
Chapter Test 211
8. Minimize
C = 5) + 2y 2
subject to
3v, + 2y 2
y, + 4v : > 6
y, + v2 ^ 4
Set up the simplex tableau

of the dual problem. Label the
columns.
9. A simplex tableau of a standard

maximization problem is given
*l
Mathematics of Finance
Although this chapter is directed to those interested in business and eco-

nomics, nearly everyone can use the concepts of this chapter in some way in
everyday activities. Few people pay cash for all of their purchases, and
financing a car or a home has become common practice. When items are
financed, the total amount paid exceeds the price of the purchase; much of
this difference is interest. We therefore search for the lowest possible inter-
By making equal payments we amortize a debt, and
est rate. in this chapter
we learn how to find the size of these equal payments.
you are the manager of a business, these concepts are even more
If
important. Not many years ago money could be borrowed by large secure
corporations for 3% and small businesses would not pay more than 4%.
However, in recent years this rate has varied from 9% to 15%. The high cost
of borrowing money affects many management decisions today. Thus, an
understanding of this chapter could be important to you as a future manager.
In this chapter, you will be introduced to simple interest, compound
interest, geometric progressions, annuities, amortization tables, sinking
funds, perpetuities, and equations of value.
Although there are tables that perform certain financial calculations,
we prefer to use a calculator or a computer because even the most complete
set of tables is limited, and with a calculator you can handle a much wider
range of problems.
212
Some Comparisons of Interest Rates 213
4.1 Some Comparisons of Interest Rates
Overview Kate deposited $100 in a savings account paying 6% per year compounded quarterly. At the end
of the year, her statement indicated that she had $106 in her account. She believes there is a
mistake. Is she right or wrong?
The phrase compounded quarterly indicates that this problem involves compound inter-
est. What is the difference between compound and simple interest? You will learn the difference
in this section, and you will use both concepts to solve a variety of problems. In this section you
will study:
Terms associated with simple interest

A simple-interest formula relating amount and principal
The concept of compound interest
A compound-interest formula relating amount and principal
Time diagrams
The cost of borrowing money is called interest. Simple interest is computed

as a constant percentage of the money borrowed for a specific time, usually a
single year or less, and is paid at the end of the specified time.
The sum borrowed is called the principal P or sometimes the present
value; r denotes the rate of interest, usually expressed in percentage per
year; and t is time expressed in years or fractions of years. By definition,
simple interest / equals principal multiplied by the interest rate multiplied by

the time in years.
/ = Prt
EXAMPLE 1 Four hundred dollars is borrowed at 12% simple interest for 3 years. What is
the interest?
SOLUTION
/ = Prt 12% as a
decimal is 0.12
I = ($400)(0.12)(3) = $144
At the end of the term of the loan, the borrower must pay not only the
sum that was originally borrowed, the principal. Hence,
interest but also the
the amount due at the end of the term is given by
Amount = Principal + Interest
A = P + I = P + Prt
Using the distributive property, we can write this formula as
A = P(l + rt)
214 Chapter 4 Mathematics of Finance
Simple Interest l = Prt
A = P + Prt
= P(1 + rt)
where P = principal or present value, r = annual simple-interest rate, t = time in number

of years, and A = amount or future value.
EXAMPLE 2 A loan of $1000 is made for 6 months at a simple-interest rate of 8%. How
much does the borrower owe end of 6 months?
at the
solution In the formula replace P, r, and t with their values.
P = $1000
r = 0.08
1 Convert 6 months to years; fi = h. year
A = P(\ + rt) = $1000 1 + (0.08)

(|)
= $1040
The borrower owes $1040>
The relationship between principal, or present value, and amount, or

future value, for Example 2 is shown in the time diagram of Figure 1.
$1000 $1040
Principal, or Amount, or
present value future value
FIGURE 1
To find the present value of an amount at simple interest rate r, replace

A, r, and t with the given values and solve for P.
EXAMPLE 3 Compute the present value of $1000 due in 3 months with interest at 12%
annually.
SOLUTION
A = $1000
$1000 = P [l + (0.12)
(-^J r= 0.12
/
-3-
? '/ ^P = $970.8737864
' 12
In this chapter, we will round all final answers (but not intermediate steps) to
two decimal places (nearest cent). Hence, the present value is $970.87. The
time diagram for Example 3 is shown in Figure 2>
$970.87 $1000
I
I
FIGURE 2
EXAMPLE 4 One thousand dollars is borrowed for 2 years. At the end of that time, $1 120
is repaid. What percent of simple interest was charged?
A = $1120
A= P(\ + it) P = $1000
t = 2 years
$1120 = $1000(1 + 2r)
$1120 = $1000 + $2000/-
= 0.06 = 6% simple interest
$1000 $1120
I I
FIGURE 3 2 years
Let's work the preceding example another way after examining the
time diagram (Figure 3). Note between $1120 and $1000 is
that the difference
the interest.
/ = $1120 - $1000 = $120

/ = Pit
$120 = $1000(r)(2)
r = 0.06 = 6% simple interest rate
Simple interest is not always as simple as it appears. Consider the story

of Jill. Jill anticipated an IRS tax refund of $300 in early May. Being short on
cash, she decided on January 19 to borrow $300 until May 10. Her banker
agreed to give her the loan at 7.5% simple interest. First the banker calcu-
lated the number of days she would have the money.
Days remaining in January 31 - 19 = 12
Days in February 28
Days in March 31
Days in April 30
Days in May _10

111
She was comfortable with this computation, but she was quite surprised
when the banker computed her interest:
/ = Pit
1 = ($300X0.075) {)
I = $6.94
The banker pointed out that the fraction Mo indicated the portion of the year
that she would have the money. Bankers often compute interest on the basis
of a 360-day year. In the days prior to widespread use of calculators, this
convention made computations easier. Today some bankers use a 365-day
year and some a 360-day year.
Compound Interest
We have learned that simple interest / is found by using the formula / = Prt,
where P represents the principal, r the rate, and t the time. When interest is
computed by this formula, the principal always remains the same. If the
interest is added to the principal at the end of each interest period, so that
the principal is increased, the interest is said to be compounded. The sum of
the original principal and all the interest is called the compound amount, and
the difference between the compound amount and the original principal is
the compound interest. A comparison of simple and compound interest is
given in the following example.
EXAMPLE 5 Find the simple interest on $1000 for 3 years at 6%. Then find the compound
interest on $1000 for 3 years at 6%, compounded annually.
solution First we summarize the problem on a time diagram (see Figure 4).
First Second A (at end

interest interest of third
P period period period
I I I 1
FIGURE 4 I 2 3 years
To emphasize the difference between simple and compound interest, we will
compute both year by year. At the end of the first year, the interest for both
will be
/ = $1000(0.06) = $60
For compound interest, this amount is added to the principal, and the new
principal becomes $1000 + $60 = $1060. Interest for the second year is
/ = $1060(0.06) = $63.60
Now, how do you get the principal for the third year?
TABLE 7
TABLE 2
Period
solution r = 0.12, k = 4, t = 5, and n = kt = 4 5 = 20. Thus,
= $2000(1 + 0.03)
20
= $3612.22
The compound interest is
$3612.22 - 2000 = $1612.22
Once we have established the relationship A = P[l + (r/k)] kl we can ,
use this formula to change the direction of our thinking. Consider the ques-
tion, "How much principal must we invest now at 8% per year compounded
quarterly in order to have the $6000 we need to buy a used car in 4 years?"
When the question is asked in this way, the principal for which we are
searching is called the present value. Thus, we are looking for the present
value of the compound amount $6000 due in 4 years at 8% per year com-
pounded quarterly.
J
FIGURE 1 4 years
A = $6000
A = P[l r = 0.08
+l )
=4
k
$6000 = P(\ + 0.02) 16 n = kt= (4)(4) = 16
$6000
P = 16
(1.02)
CALCULATOR NOTE Do you have a reciprocal key [

l/.v |
on your calculator? Try
1.02 [71 16 Fl \UP[ [^\ 6000

to get $4370.67. Thus, the present value of $6000 due in 4 years at 8%
compounded quarterly is $4370.67.
EXAMPLE 8 How much money should be deposited in a savings and loan association
6% compounded quarterly in order to
paying have $3000 in 5 years?
$3000
FIGURE 8 5 years
220 Chapter 4 Mathematics ot Finance
SOLUTION
A = P\l +
i = .06/4 = 0.015
n = kt = = 20
(4)(5)
$3000 = P{\ + 0.0 15) 20
$3000
(1.015) 20
Using a calculator, we find
P = $2227.41
Thus, $2227.41 must be deposited now in a savings and loan association in

order to have $3000 in 5 years.
EXAMPLE 9 How long will it take a dollar to double at 8% compounded semiannually?
solution Suppose we start with P= $1. This amount doubles or A = $2. r is
0.08 and A = 2 (semiannually). Substitute these values in
2 = (1.04)"
CALCULATOR NOTE Let's use the |y' key of your calculator and the very useful mathematical
|
tool of "guessing and then checking" to get an approximate answer. Try any
number for /;. Suppose you select a 5.
1.040 5=1.217 1
Much too small]
1.04 010=1.480 |
Still too small]
1.04 020 = 2.191 |

A bit too large]
1.04 17 = 1.948
1.04 18 = 2.026
Thus, in 18 interest periods (or 9 years) the principal will have more than
doubled.
Practice Problems i> 1. Steve Jones borrows $5000 to complete his senior year at Roebuck Col-
lege. If the bank charges 8% simple interest, how much will Steve owe in
3 years?
2. In 5 years, you will need $10,000 as a down payment on a condominium.
How much must you deposit today if the Secor Savings and Loan will pay
you 8% simple interest?
3. You borrow $1000 and the Easy Loan Company requires that you pay
$1240 in 2 years. Which simple-interest rate is the loan company charg-
ing?
4. You want to borrow $10,000 for 6 years. What is a better interest rate:
(a) 8j% simple interest or
(b) 8% compounded quarterly?
Answers to practice problems are found on p. 223.
Exercise Set 4.1 1. Compute the simple interest (b) $3000 at 6% compounded
when the principal, rate, and semiannually for 6 years
time of the loan are given. (c) $3000 at 6% compounded
(a) P = $500, r = 0.08, t = 2 quarterly for 6 years
years
7. Compute the amount to be re-
(b) P = $300, r = 0.13, t = 4
paid when the principal, simple-
years
interest rate, and time of the
(c) P = $500, r = 0.10, t = 5
loan are given.
years
(a) P = $4000, r = 0.11, t = 2
2. What is the amount to be repaid years
in (a), (b), and (c) of Exercise 1? (b) P = $3000, r = 0.12, t = 6
months
3. Find the interest and the amount
(c) P = $100, r = 0.08, t = 3
of a loan for $3000 borrowed for
years
2 years at 16% simple interest.
8. Find the simple-interest rate on
Find the compound interest and
the loan when the principal, the
compound amount for the invest-
amount repaid, and the term of
ments in Exercises 4 through 6.
the loan are given. Hint: Use a
4. (a) $5000 at 12% compounded 360-day year.
annually for 3 years (a) Principal, $3500; amount re-
(b) $5000 at 12% compounded paid in 2 years, $4130.
semiannually for 3 years (b) Principal, $1000; amount re-
(c) $5000 at 12% compounded paid in 120 days, $1046.67.

quarterly for 3 years (c) Principal, $500; amount re-
paid in 45 days, $510.
5. (a) $2000 at 10% compounded
annually for 4 years Find the present value of the money
(b) $2000 at 10% compounded in Exercises 9 through 12.
semiannually for 4 years
9. $5000 due in 5 years if money is
(c) $2000 at 10% compounded
worth 16% compounded annu-
quarterly for 4 years
ally
6. (a) $3000 at 6% compounded
annually for 6 years
10. $6000 due in 5 years 3 months if Solve each formula for the indicated
money is worth 8% compounded variable.
semiannually
13. / = Pit (r)
11. $7000 due in 4 years if money is

14. A = P + Prt (P)
worth 12% compounded semian-
nually 15. How long will it take for $125 to
amount to $375 at 12% interest
12. $8000 due in 6.5 years if money
compounded quarterly?
is worth 7% compounded annu-
ally 16. How many years will it take to

double $1000 at 16% interest
compounded semiannually?
Applications
(Business and 17. Interest. Find the interest on 22. Loans. Kate borrowed $700
Economics) $2000 borrowed for 8 months at and agreed to repay the principal
10% simple interest. with interest at 8% compounded
Find the amount of a
semiannually. What will she
18. Interest.
90-day, $1500 loan at 12% sim-
owe at the end of 5 years?
ple (Use a banker's

interest. 23. Loans. On April 1, 1986, Rana
year of 360 days in the computa- borrowed $3000 at 8% com-
tion.) pounded quarterly. What will
19. Interest. How much interest she owe on October 1, 1998?
will you owe on a $1000 loan 24. Savings. The sum of $1000 was
from March 3 to August 7 at 10% deposited in a bank at an interest
simple interest? (Use a 365-day
rate of 6% compounded semian-
year.)
nually. Five years later the rate
20. Interest. What is the interest increased to 8% compounded
on a $1500 loan from June 15 to semiannually. If the money was
September 11 at a simple inter- not withdrawn, how much was
est rate of 8%? (Use a 365-day in the account at the end of 6
year.) years?
21. Loans. Some lending institu- 25. Savings. How much should
tions have a minimum amount of parents invest for their daughter
interest they must collect on any at 12% interest compounded
loan. An institution with a mini- semiannually in order to have
mum charge of $5 will expect to $5000 at the end of 20 years?
receive $5 interest on the loan,
26. Cash Value. A lot is sold for
even though actual interest
$750 cash and $600 a year for the
charges are only $3. How long
next 3 years. Find the cash value
must $1000 be borrowed at 8%
of the lot if money is worth 6%
interest to reach the minimum
compounded semiannually.
service charge of $5? (Use a
banker's year of 360 days in the
computation.)
Effective Rates, Continuous Compounding, and Geometric Progressions 223
and Life Sciences) The effects of inflation can be ob- (a) 3%
tained using the formula for com- (b) 4%
pound interest where i becomes the (c) 5%
inflation rate per period. (d) 6%
27. Inflation. Find the cost of the 31. Population Growth. The popu-
following items in 10 years at an lation of a city of 60,000 is ex-
annual inflation rate of 6%. pected to increase at a rate of

(a) A $40,000 house 4% per year for the next 10
(b) A 900 hamburger years. What is the population at
(c) A $3.50 movie ticket the end of 10 years?
(d) A $14.65 hourly labor rate. 32. Fish Population. The number of
28. Inflation. Rework Exercise 27 fish in a lake expected to in-
is
for an 8% annual inflation rate. crease at a rate of 12% per year

for 5 years. How many fish will
29. Inflation. Rework Exercise 27 be in the lake in 5 years if 10,000
for a 10% annual inflation rate.
are placed in the lake today?
30. Inflation. Find how long it will
take a price to double with an

average annual inflation rate of
Answers to > 1. A = P(\ + rt), or A = 5000(1 + 0.08 3) = $6200.
Practice Problems 2. A = P(l + rt), or 10,000 = P{\ + .08 5), or 10,000 = P(\A0): P =

$7142.86.
3. / = 1240 - 1000 = 240. Then / = Pit, or 240 = 1000(f)(2), or 240 = 2000/-.
f = 0.12 = 12%.
4. (a) A = 10,000(1 + 0.0825 6) = $14,950.

(b) A = 10,000(1 + 0.02) 64 = $16,084.37. The loan at 8i% simple interest

is the better rate.
4.2 Effective Rates, Continuous Compounding, and

Geometric Progressions
Overview Effective rates and continuously compounded rates are useful extensions of our discussion on
compound interest. Geometric progressions are introduced to facilitate an understanding of the
next section on annuities. In this section we consider the following:
A formula for effective rates
A formula for continuous compounding

The definition of a geometric progression
A formula for the sum of a geometric progression
We have not yet established how to compare the interest rates prom-
ised by two institutions. For example, if the bank down the street offers an
interest rate of10% compounded 5 times a year and the bank 12 miles across
town 12% compounded 3 times a year, should we undertake
offers a rate of
the long drive? To compare the two rates, we need to introduce the notion of
an effective rate of interest. The effective rate of /% per period compounded
k times a year is the simple-interest rate that gives the same amount due at
the end of year as the rate per period compounded k times a year. That is,
1 i
P(l r 1) = P(l + 0*
I + r = (1 + if Divide by P
r = (1 + 0* - 1
where / is the simple-interest rate, and i is compounded k times a year. From

this relationship the effective rate can be obtained or the following formula
can be used:
Definition: Effective A rate of / per period compounded k times a year produces an effective annual rate of
Annual Rate
Effective rate = (1 + /)* - 1 / = rlk
EXAMPLE 10 Find the effective rate equivalent to a nominal rate of 6% compounded

quarterly.
SOLUTION
Effective rate = (1 + 0.015)
4 - 1
= 1.0614 - 1
i =-3* = 0.015
= 0.0614 k = 4
Thus, a nominal rate of 6% compounded quarterly is equivalent to an effec-

tive rate of 6.14%>
EXAMPLE 11 For a savings account, which is the better rate, 12.5% compounded semian-
nually, or 12% compounded monthly?
solution To compare the two rates, we first find and compare the effective
rates:
(1 + .0625) 2 - 1 = .128906 or 12.8906%
(1 + 0.01) 12 - 1 = .126825 or 12.6825%
The effective rate for 12.5% compounded 2 times a year is greater than 12%
compounded 12 times a year>
A comparison of some compound-interest formulas follows:
Summary of Compound Amount at compound interest
Interest
Interest
$8.50 T
Interest on $100 invested
for 1 year at 8% compounded
as shown
$8.00
Annually Semiannually Quarterly Monthly Daily Continuously
FIGURE 9
CALCULATOR NOTE If there is no \7\ or [e^\ button on your calculator, you can usually obtain the
value of e 3 by
3 |
INV |
[TnT] [^"1
3 1
2nd |
fln7| f=1
or
2.71828 7 3 [=1
[y ]
In Figure 9, note the rather significant increase in interest earned as the

number of compound-interest periods per year changes from 1 to 2 to 4 to 12.
Then note the insignificant change in interest earned from compounding
daily to compounding continuously.
EXAMPLE 12 Find the compound amount obtained from an investment of $2000 com-
pounded continuously for 5 years at 6%.
SOLUTION A = Pe" = 2000e 5(0 06
= 2000? 03 = $2699.72
CALCULATOR NOTE Try 5 [x] .06 1

2nd I
j
ln-v |
[ <] 2000 [=] The answer is $2699.72.
Geometric Progressions
The next two sections involve a sequence of payments or deposits that draw
interest. In the preceding section, we learned that to accumulate a payment
with compound interest we multiply by the factor (1 + /)". To find the sum of
such terms, we introduce what is known as a geometric progression.
A geometric progression is characterized by the fact that each term is
obtained from the previous term by multiplying by the constant r, called the
common ratio. Thus, the sum of a geometric progression can be written as
~
S = a + ar + ar 2 + ar 3 +
+ ar n x
where r is the common ratio.
EXAMPLE 13 Determine whether the following is a geometric progression, find the first
term, and find the common ratio.
12 + 36 + 108 + 324 +

SOLUTION The first term is a = 12. Since 36 = 12 3, 108 = 36 3 = 12 3 2

,
and 324 = 108

3 = 12
3
3
, this expression represents a geometric progres-
sion with r = 3>
The sum of the first n terms of a geometric progression can be ex-

pressed as
S = a + ar + ar 2 + + ar"' 1
1st 3rd 2nd nth

term term term term
Let's multiply S by r and subtract rS n from S.
S n = a + ar + ar 2 + + ar"~ l
rS n = ar + ar 2 + + ar" '
+ ar"
S n rS n = a ar"
(1 - r)S = a(\ - r
n
) Factor out an a and S
If r 1, dividing by 1 - r gives
a(\ - r") a(r" - 1)

Sn =
Thus, we have the following:
Sum of a Geometric The sum S n of the first n terms of a geometric progression with first term a and common
^
Progression ratio r + 1 is
a(1 - r")
Sn = or Sn =
1 -r
EXAMPLE 14 Find the sum of six terms of the following geometric progression:
6 + 12 + 24 +

SOLUTION Since this is a geometric progression, each term is obtained from

the preceding term by multiplying by a constant r. The constant r may be
found by dividing any term by the preceding term
r = 126 = 2 or r =
24
= 2
11
After r is found, other terms of the progression may be obtained by multiply-

ing the preceding term by r. The sum of n terms is found by using the formula
a(r" - 1)
5 =
r - 1
2 - 1
= 378
EXAMPLE 15 Find the sum of
100 + 100(1.06) + 100U.06) 2 + + 100(1.06)

4
SOLUTION Notice that S is now the sum of five terms of a geometric progres-
sion whose first term is 100 and whose common ratio is 1.06; hence,
_ a(r" - 1) _ 100[(1.06) 5 - 1]
*5
r-\ 1.06-1
% CAWLATORNOTE Use 1 .06 Fl 5 Fl F] 1 f^l |"*1 100 FH Rl 1-06 F1 1 [71 F1

or 1OOE01-O6H5B10HH1-O6B1EE]
5s = 563.71
EXAMPLE 16 Find the sum of
100(1.06)"' + 100(1. 06)"

2
+ 100O.06)- 3 + 100(1.06)-" + 100(1.06)- 5
SOLUTION First write this with positive exponents:
100 100 100 100 100

+ TT-^32 + 4
(1.06) 5
3
(1.06)' (1.06) (1.06) (1.06)
Then note that this is a geometric progression with a = and common

1.06
ratio

1.06
. To verify, multiply the first term by and see
1.06
if you get the second
term, and so on through the progession. Thus, the sum is

100
1.06 L (1.06)
5 Use
1 (1 - /")
i, = a
1.06
,
1 - r
100 1 - (1.06)
1.06 1 .06 - 1
1.06
1 - (1.06)" 5
= 100
0.06
= 421.24
Using the formula for the sum of a geometric progression gives
1
Use
Sn = a(l - r")
S
1 - r
YTl) =
(l + o (i - (l + - 1
Practice Problems 1. Find the effective rate equivalent to 8% compounded monthly.

2. To what amount will $10,000 accumulate in 6 years if the 8% interest is
compounded continuously?
Answers to practice problems are found on page 231
Exercise Set 4.2 Identify whether the following are 5. a, a + r, a + 2r, a + 3r, .
geometric progressions. If so, find

2 3
r r r
the common ratio.
1
'
k'k'T'T' ' '
1. 4, 12, 36, 108, . . .
2
7. a, a(l + , ad + 0\
2. 6, 14, 22, 38, 46, . . .
-1
8. a, o(l + , fld + 0" 2 ,
3. 12, 9, 6, 3, 0, -3,
a(l + 0"'i
3 3
4. 12, 3, -r.TT,
4 16
Applications
(Business and Find the effective rates equivalent to (d) Monthly
Economics) the nominal rates for Exercises 9- (e) Daily
12. (f) Continuously
What is the amount after 10
9. 8% compounded semiannually
years?
10. 6% compounded semiannually
14. Find the compound amount for
11. 24% compounded monthly (a) $3000 at 12% compounded
continuously for 5 years
12. 12% compounded monthly
(b) $6000 at 15% compounded
13. $1000 is invested at 12% com- continuously for 6 years
pounded
15. Find the effective rate corre-
(a) Annually
sponding to 10% interest com-
(b) Semiannually
pounded continuously. (Hint:
(c) Quarterly
Amount of an Annuity: Sinking Funds 231
You must adjust the formula 18. How many years will it take
given.) money to double if 8% interest is
compounded daily?
16. Approximately how many years
will it take money to double at 19. What is the effective rate of 8%
10% interest compounded con- compounded daily?
tinuously?
17. Find the accumulated value of

$10,000 invested for 10 years at
12% interest compounded daily.
and Life Sciences) 20. Fish Population. Suppose the many fish should the lake con-
instantaneous rate of growth of tain 6 years from today? (Hint:
the fish population in a lake is AT = I00e l0
-
Find N and
>.
10% per year. If you stock the round to the nearest whole num-
lake with 100 fish today, how ber.)
Answers to Practice C> 1. r = (l + = .083 or 8.3%

j^J
Problems
2. A = $10,000e 6(008) = $16,160.74
4.3 Amount of an Annuity: Sinking Funds
Overview In this section we are interested in a sequence of equal payments made at the ends of equal time
intervals. Deposits (of equal amounts) in a savings account at the end of each quarter fit our
pattern. For example, suppose you deposit $500 at the end of each quarter in a savings and loan
that pays interest at 8% compounded quarterly. You might wish to know if you have enough
money in this savings account at the end of 5 years to purchase the automobile of your dreams.
At the end of this section we will investigate how money accumulates by studying sinking funds.
An
annuity is a sequence of equal payments made at equal time inter-
vals. For the formula we will develop, these payments, denoted as R, are
made at the ends of equal successive payment periods, and the interest
periods are the same as the payment periods. The sum of all payments R,
plus their interest, is called the amount of an annuity as illustrated in the
following example:
EXAMPLE 19 Suppose at the end of each year you receive $100 and invest it at 6% com-
pounded annually. How much would you have at the end of 5 years?
SOLUTION Let 5 represent the amount of the annuity at the end of 5 years.
Since the firstpayment of $100 is not received until the end of the year, it will
accumulate interest for 4 years as seen in Figure 10. Likewise, the second
payment will accumulate interest for 3 years, and so on. Hence,
$5 = $100(1.06)" + $100(1.06) 3 + $100(1.06) 2 + $100(1.06) + 100
$100(1.06)
4
$100(1.06)
3
$100(1.06)
2
$100(1.06)
$100
B100 $100 $100 $100 $100

I I I I I
FIGURE 10
Using the preceding example as a pattern, we now derive an expres-

sion for the amount of an annuity of $R deposits at the end of intervals for n
intervals as shown in Figure 1 1
$/?(i +/)"-'
7
$/?d +i)"'
3
$/?(l + i)
2
$(1 +/)
$/?0 +i)'
$/?
$R $/? $/?
I I I
FIGURE 11
$S = $R + $R(l + i) + $R(l + i)
2
+ $^(1 + if
+ +$/?(! + i)"-
2
+ $R(l + /)"-
Using the formula for the sum of a geometric progression in Example

17, page 229, this sum can be expressed as
(l + 0" - i
S = R
This is summarized as follows:
Amount of an Annuity (1 + /')" - 1
S = R
where S = amount of the annuity, fl = periodic payment of the annuity, / = rate per
period, and n = number of payments (periods). (Payments are made at the end of each
period.)
(1 + /)" - 1
is sometimes abbreviated s^,-.
The use of the formula is illustrated by the following example.
EXAMPLE 20 Kate deposits $300 at the end of each year in a savings account that pays 8%
interest compounded annually. How much money does she have just after
the fifth deposit?
5
$300 $300 $300 $300 $300
I _| I I I
FIGURE 12 4 5
SOLUTION Remember that S gives the amount just after a payment is made.
Hence, it can be used to solve the problem.
((I + i) - V i = 0.08
S = R n = 5
R = $300
((I + 0.08)
5
- 1'
S = $300
V 0.08
= $1759.98
Thus, Kate has $1759.98 after the fifth deposit.
EXAMPLE 21 Mark deposits $1000 in the First Savings and Loan at the end of each quarter
for 10 years. How much money does he have at the end of 10 years if the
savings and loan pays 12% interest compounded quarterly?
5
$1000 $1000 $1000 $1000 61000
I I I I _l
FIGURE 13 10 years
SOLUTION
'd + /)"- r
S = R /
=
4
(1 + .03)
= $1000
.()}
= $75,401.26
EXAMPLE 23 Mr. Wilkins wishes to create a sinking fund to pay off his loan, which will
amount to $1259.71 in 3 years. If his sinking fund pays 12% interest com-
pounded semiannually, what is his semiannual deposit into the sinking fund?
solution Deposits of $R in the sinking fund are shown in Figure 15.
$1259.71
$R $R $R
_l_
FIGURE 15 3 years
(1 + i)" - 1 / = rlk = 0.12/2 = 0.06

S = %R = = 2(3) = 6
n kt
5 = $1259.71
(1 + 0.06) 6 - 1
$1259.71 = $R
0.06
6
CALCULATOR NOTE Compute (1.06) ; subtract I; divide by 0.06; take reciprocal; multiply by
1259.71.
R = $180.60
A schedule for the fund in the preceding example is given in Table 3.
TABLE 3
The second $180.60

$180.60 + $10.84 + $180.60 = $372.04
is the new deposit.
Now can you verify the other entries in Table 3?
Practice Problems > 1 . Ifyou deposit $100 a month in a savings and loan paying 6% compounded
monthly, how much money do you have at the end of 5 years?
2. A self-employed person is setting up a defined-benefit retirement plan
with a goal of having $100,000 in the plan at the end of 5 years. What
amount must be deposited in the fund each month if the fund accumulates
interest at 9% compounded monthly?
Exercise Set 4.5 With a calculator, compute each of 9. R = $2000, r = 12% com-
the following. Interpret what you pounded monthly, n = 4 years
have found.
10. R = = 14.5% com-
$5000, r
- '
pounded monthly, n = 10 years

LlOO^ 0.06
1
11. Find the amount of the following

(1.01)
100 - 1 annuities.
2. 50
0.01 (a) $1000 per year for 20 years
at 8% interest compounded
/(1.08)
14
- 1
annually
3. 200
0.08 (b) $500 per quarter for 6 years
(1.12)
17
- V at 8.5% compounded quar-
4. 1
terly
0.12 /
(c) $600 per half year for 5 years

Find the amount of the following an- at 8% compounded semian-
nual annuities. Interest is com- nually
pounded annually. 12. Exam. A businessperson wants
5. R = $100, i = 6%, n = 10
to invest a certain sum of money
at the end of each year for 6
6. R = $1000, i = 8%, n = 12 years. The investment will earn
Find the amount of each of the fol-
compounded annually. At
the end of 10 years, the busi-
lowing annuities. Payments or de-
posits aremade at the end of each nessperson will need $100,000.
interest period.
How should the person compute
the required annual investment
1. R = $100, r = 6% compounded
semiannually, n = 10 years
8. R = $1000, r = 9% compounded
quarterly, n = 5 years
$100.000 (1.06)
(a) (c) $100,000
1
- (1.06) 0.06
0.06
1
- (1.06)
(d) $100,000
$100,000 0.06
(b)
(1.06)'" - 1
0.06
Applications
(Business and 13. Investments. Juan is to receive 16. Sinking Fund. Compute the
Economics) $1000 at theend of each year for quarterly deposit Mr. I. O.
5 years. If he invests each year's Yeung must make to a sinking
payment at 8% compounded an- fund that pays 12% interest com-
nually, how much will he have at pounded quarterly to pay off a
the end of his 5 years? loan of $5000 due in 4 years.
Construct three lines of a sched-
14. Investments. Suppose you de-
ule for this fund.
posit $500 each 6 months in a
credit union that pays 8% inter- 17. Sinking Fund. What deposit
estcompounded semiannually. must be made to a sinking fund
How much would you have after that pays 12% interest com-
5 years? pounded quarterly to pay off a
loan of $1500, due in 4 years, at
15. Retirement Account. John Size-
more deposited $2000 per year
8% interest compounded annu-
ally? Construct three lines of a
in a retirement account. Make a
schedule for this fund.
table showing how his money
accumulates for the first 5 years 18. Sinking Fund. What deposit
and then find how much he will must be made to a sinking fund
have at the end of 20 years if his that pays 8% compounded quar-
bank pays terly to pay off in 5 years a loan
(a) 8% compounded annually of $3200 at 12% interest com-
(b) 10% compounded annually pounded semiannually?
and Life Sciences) 19. Inflation. If inflation holds 20. Fish Population. You place
steady at 4.2% per year for 5 1000 fish in your lake each year
years, what will be the cost in 5 for 5 years. If the fish increase at
years of a $10,000 car today? a rate of 5% per year, how many
How much will you need to de- will you have at the end of 5
posit each year in a sinking fund years?
earning 8% per year to purchase
the new car in 5 years?
Answers to > (1.005)

60 - V
1. $100 $6977
Practice Problems ( .005
'( 1.0075)"
2. $R = $100,000
.0075
$R = $1325.84
4.4 Present Value of an Annuity and Amortization
Overview In the previous section we found the value of periodic deposits just after the last deposit. Now
we are interested in finding the value of the periodic deposits one payment period before the
first deposit. For example, we need to know how much money must be invested now in order to
provide periodic payments for a number of periods in the future. In this section, we will
Define the present value of an annuity

Develop a formula for the present value of an annuity
Find regular payments to amortize a debt

Show what happens with each payment by constructing an amortization schedule
EXAMPLE 24 Compute the present value of an annuity of $100 per year for 5 years at 6%
compounded annually. In other words, find the amount of money that must
be invested now at 6% compounded annually so that payments of $100 per
year can be made from this investment for 5 years.
SOLUTION In Figure 16, $100(1.06)"' must be invested now

have $100 in 5
to
4
years. That is, 100(1.06)"' (1.06)' = 100(1.06)= must be
100. $100(1.06)
invested now to have $100 in 4 years. Finally, $100(1.06)"' must be invested
now to have $100 in year. The present value of the annuity
1 is the sum of
these terms.
$100(1 +0.06)" 5
Present Value of an Annuity and Amortization 239
Thus,
$P = $100(1.06)-' + $100(1.06) :
+ $100(1.06)- 3
4
+ $100(1.06) + $100<I.06r s
$100 $100 $100 $100 $100

$P =
I.06) : 3 4
(1.06) 5
1
(1.06) (1.06) (1.06)
Before finding the value of the preceding expression, let's follow the
same pattern for a general case. Consider $R payments at the ends of pay-
ment intervals for /; intervals. The rate per interest period is The value at /'.
/ = of the first $R is R(] + /)'. The second $R has a value at / = of

R{1 + i) : The last $/? payment has a value of R(\ + /)" at t = 0.
.
$m\ + i)
EXAMPLE 25 Compute the present value of an annuity that pays $100 each month for 4
years at 12% interest compounded monthly.
SOLUTION
k= 12
$100 $100 $100 $100 $100 $100

I I I I I I
FIGURE 18 3 '
4 years
i = rlk = 0.12/12 = 0.01
n = kt = 12
4 = 48
(1 - oir '
= $ioo[i--
0.01 R = $100
7 48
|j|
CALCULATOR NOTE Compute 1.01 [y ] [] [^] [] [+] 1
P] [T| .01 fx] 100 p]
P = $3797.40
EXAMPLE 26 Kamilla purchased a refrigerator for $150 down and $30 a month for 12
months. If the interest charge is 12% compounded monthly, find the cash
price.
payment less interest ($88.85 - $10 = $78.85) gives that part of the payment
that is applied to the debt. So, at the beginning of the second month, the debt
is $1000 - $78.85 = $921.15. Can you find this on the first line of Table 4?
Now see if you understand the second line.
Practice Problems > 1, Jodi agrees to pay $500 down, $200 per month for 5 years for her new
automobile. If the finance rate is 12% compounded monthly, what is the
cash value of her car?
2. You purchase a house for $100,000, pay 20% down and amortize your
debt with monthly payments for 30 years. What is your monthly payment
if your loan charges 9% compounded monthly?
3. Show three lines of the amortization schedule in Practice Problem 2. That
is, show how the payments contain both interest and payment on the
principal.
Exercise Sel 4.4 With a calculator, compute each of (b) $500 per quarter for 6j years
the following. Interpret what you at 8.5% interest com-
have found. pounded quarterly
(c) $600 per half year for 5 years
1
- (1.06)-
1. 50 at 8% interest compounded
0.06
semiannually
1 - (1.01)-
2. 50 Find the present value of the fol-
0.01
lowing annuities:
- (1.08)-' 4 (a) $100 per month for 8 years at
3. 200 12% interest compounded
0.08
monthly
(1.12)- 7 \
$500 per month for 3 years at
-
(b)
4. 1
0.12 / 12.4% interest compounded

Find the present value of each of the monthly
following annuities. Interest is com- Exam. A businessperson wants
pounded annually. to receive $6000 (including prin-
cipal) from a fund at the end of
5. R = $100, / = 6%, n = 10
each year for 10 years. How
6. R = $1000, i = 8%, n = 12 should she compute her required
initial investment at the begin-
7. Find the present value of the fol-
ning of the first year if the fund
lowing annuities:
earns 6% compounded annu-
(a) $1000 per year for 20 years
ally'.'
annually
6000 6000
(a) (c)
(1.06)' (1 - (1.06)-
0.06 0.06
(1.06)
10 - 1 1
- (1.06)
(b) 6000 (d) 6000
0.06 0.06
Applications
(Business and 10. Payments. Compute the month- (a) What is her equity after 30
Economics) ly payment necessary to finance payments? (Hint; $60,000
a used car for $3500 at 12% in- less the present value of a
terest compounded monthly for 66-payment annuity)

3 years. (b) What is her equity after 60
payments?
11. Payments. Find the payment
necessary each quarter for 2 15. Equity. Steven bought a cabin
years to amortize a debt of $2000 He paid $5000 down
for $32,000.
at 12% interest compounded and amortized the balance with

quarterly. monthly payments for 7 years
12. Amortization. Make an amorti- monthly.
zation schedule for Exercise 10. (a) What is his equity after
13. Amortization. Make an amorti- his 50th payment? (Hint:

$32,000 less the present
zation schedule for Exercise 11.
value of a 34-payment an-
14. Equity. Pai Ling
bought a nuity.)
house for She paid
$60,000. (b) What is his equity after his
$10,000 down and amortized the 80th payment?
balance with monthly payments
for 8 years at 12% interest com-
pounded monthly.
Answers to Practice 1. Cash value = $500 + Present value of a $200 annuity

Problems
ri -(1 + o.OD-
~
= 50 + 20
i oIm
= 500 + 8991.01
= $9491.01
2. $100,000 - 0.20($100,000) = $80,000 (owe on the house)
1 - (1 + 0.0075)- 360 ]
J
$80,000 *[ 0.0075
R = $643.70
3.
Outstanding Principal
Month Principal Interest Due Payment Repaid
1
Equations of Value and Perpetuities 245
$/? $( + /)"
12
I
I I I I I
(a)
3 n
$R = + iY"
$/?(!
(1 + /)"
$
(b) I I L
$R $R $R $R
I I I I I
(c)
FIGURE 27 o 1
The preceding ideas will be used in what we will call equations of value.
In an equation of value, you select a given date. Then you accumulate
(multiply by ( 1 + /)") or discount (multiply by ( I + /)") all obligations to this
date, as studied in Section 4.1. Likewise, you accumulate or discount all
payments to this date. When you set equal the value of the obligations and
the value of the payments on this date, you have an equation of value.
EXAMPLE 28 Ari owes $10,000. He wants to pay this debt by making payments of $1000
each end of each year for 6 years. Then, at the end of 6 years, he will
at the
make a payment of $x to retire the remaining debt. If money is worth 8%
interest compounded annually, what is the size of the $jt payment Ari will
make at the end of 6 years?
$10,000 $1000 $1000 $1000 $1000 $1000 $1000

I I I I I I I
6 years
FIGURE 22
solution We choose the 6-year mark as the time to compare the debt and
the payments. At the end of 6 years the debt has a value of $10,000(1 +
6
0.08) . At the end of 6 years the $1000-a-year annuity has a value of
$1000
(1
'qq
~ *
Sum of the annuity
| |
[ 8 ]
Of course, x has a value of x at the 6-year mark. Thus, the equation of value
at the 6-year mark is
{L ] l
10,000(1 + 0.08)
6
= 1000 Q + x Debts equal
]
[ 0S payments at
x = $8532.81 the 6-year
mark
Thus, the one additional payment at the end of 6 years that will retire
the debt is $8532. 81.
EXAMPLE 29 Use an equation of value at time to find x for Example 28.
SOLUTION Debt = Present value of an annuity + Present value of payment at
6-year mark
8)
10,000 = iooo + *n.08)- 6
P ~oo8 ]
: CALCULATOR NOTE 1 .08 |

y '
|
6 [| ^] (store answer on a piece of paper or in your calculator)
LULtllHIZI 1000|T]-08E][i][] 10-000 Fl Fl (answer stored) Fl
x = $8,532.81
As expected, the value of x is the same regardless of the time selected

for the equation of value.
An annuity whose first payment is made some time after the end of the
first period is called a deferred annuity.
EXAMPLE 30 A car was bought on January 1 with the agreement that there would be 36
monthly payments of $100, the first of which would be due on April 1. Find
the equivalent cash price if interest is 12% compounded monthly.
P $100 $100 $100

I I I I I
Jan. 1 Mar. 1 Apr. 1 May 1

36 monthly
payments
I
SOLUTION Let's select March 1 as the comparison date for an equation of

value.
Pd.OD^lOO p-y 6
]
P must be
accumulated for
P = $2951.43 two periods.
The equivalent cash price is $2951. 43.

Equations of value are useful when the payment and interest periods
do not exactly fit our definition of an annuity.
EXAMPLE 31 Amanda pays $400 rent each month, payable in advance. What would be her
equivalent yearly rent at 12% interest compounded monthly if she paid it in
advance?
solution The $400 payment does not belong in the annuity because
at time
for an annuity you must have payment, interval then payment,
interval then
and so forth. As seen in Figure 24, we can use an equation of value including
an annuity within it. At time 0, the equation of value can be written as
- (1 - lr
1
Without the first
P = 400 + 400(
0.01 $400, there are only
1 1 payments of $400
$4547.05
in the annuity.
p
$400 $400 $400
I
I |__
FIGURE 24 1 2 11 months
We conclude this section with a discussion of perpetuities. An annuity

whose payments begin on a certain date and continue indefinitely is called a
perpetuity. Since the payments continue indefinitely, it would be impossible
to compute the amount of a perpetuity; however, the present value can be

found. The present value P of a perpetuity that is payable at the end of each
interest period at i% per period is the principal that would, in one interest
period, earn the payment R. Thus,
Pi = R or P=-
i
dividends of a stock remained constant and the value of a stock

If the
remained constant, a stock would be a perpetuity.
EXAMPLE 32 The Cenco Company is expected to pay $3.00 every 6 months on a share of
its stock. What is the present value of this stock if money is worth 6%
interest compounded semiannually?
S0LUT.0N R = $3M =
p = $10Q
Often, the payment periods and the interest periods for a perpetuity
may not be the same. When this happens, the present value may be found by
the following formula.
Present Value of The present value of a perpetuity that yields the payment R at the end of n interest
Perpetuity periods with money worth /'% per interest period is
(1 + /)" - 1
Notice that the preceding formula is P Rlr where r = (1 + /)" - 1 is
the effective rate for the payment period.
EXAMPLE 33 Find the present value of the Cenco Company's stock, which is expected to
pay $3.00 every 6 months, if money is worth 8% interest compounded quar-
terly.
SOLUTION Since money is worth 8% interest compounded quarterly, the in-

terest is 2% each quarter. The payment of $3.00 is made after two interest
periods; therefore, n = 2. Substituting in the equation gives
R
P = -
(1 + /)" 1
$3.00
(1.02)
2
- 1
= $74.26
Practice Problems !> 1. To pay for his car, Aaron is to give his brother $1000 at the end of each
year for 6 years. Aaron make the first two payments. At the end of
fails to
3 years, Aaron is going to borrow money to pay off his debt to his brother.
If it is agreed that he should pay interest at 8% compounded annually,
how much should Aaron pay his brother at the end of 3 years?
2. Mr. Smith plans to retire in 1 year. He wants to place in a trust fund,
bearing interest at 9% compounded semiannually, enough money to re-
ceive $9000 every year forever (starting 1 year from today). How much
money must he place in the bank?t>
Exercise Set 4.5 Find the present value of the follow- 3. $10,000 a year, interest at 8%
ing perpetuities: quarterly
1. $10,000 a year, interest at 8% 4. $20,000 a year, interest at 10%

annually semiannually
2. $20,000 a year, interest at 10% Find the payments of the following

annually perpetuities:
Present value $200,000, annual Present value $200,000, annual

payments, interest at 8% annu- payments, interest at 8% com-
ally pounded quarterly
Present value $1,000,000, Present value $1,000,000, semi-

monthly payments, interest at annual payments, interest at
12% monthly 12% monthly
Applications
(Business and 9. Stock Dividend. The Go Far If money is worth 12% inter-
Economics) Company is expected to pay est compounded semiannually,
$6.00 every 6 months on a share what will be the amount of the
of its stock. What is the present last payment?
value of this stock if money is

14. Cash Payment of Rent. Dora
worth 8% interest compounded
pays $300 rent each month, pay-
semiannually?
able in advance. What would be
10. Cash Premium. The annual pre- her equivalent yearly rent at
mium for a 20-year-pay insur- 12% interest compounded
ance policy is $100 payable at monthly if she paid it in ad-
the beginning of each year for 20 vance?
years. What is the equivalent
Payment. Find
15. Deferred the
cash premium ifmoney is worth present value of an annuity of
6% interest compounded annu- $1000 per year with the first pay-
ally?
ment due 3 years from now and
11. Cash Premium. Find the equiv- the last occurring 12 years from
alent cash premium if money is now if the interest is 8% com-
worth 8% interest compounded pounded annually.
annually for a 20-year-pay insur- 16. Equations of Value. To cancel
ance policy that has an annual three loans of $2000 due now,
premium of $100 payable at the $5000 due in 4 years, and
beginning of each year for 20 $10,000 due in 6 years, I agree to
years. pay $/? at the end of each year
for 10 years. If the interest rate
12. Stock Dividend. A stock pays
$0,225 each quarter. What is the
is 8% compounded annually,
what is my annual payment?
present value of the stock if
money is worth 12% interest 17. Installment Purchase. A stereo

compounded quarterly? $20 down and $30 a
set sells for
month for 12 months. What is
13. Payment of Debt. Paul has a
the cash price of the set if
debt of $8000 due in 5 years. He
wants to cancel this debt by pay-
interest is 24% compounded
monthly?
ing $3000 year from now,
1
$1000 3 years from now, and a 18. Installment Purchase. A car

last payment 7 years from now. was bought on March 1 with the
agreement that there would be money is worth 8% interest

24 monthly payments of $150, compounded quarterly?
the first of which is due on July __ ,,
1 Find the equivalent cash price
.
20 ' ***"*
, .
*?*?
,
^
TV set
that c a* h lce of $ 337 26
if interest is 12% compounded
Is s0 Jf
5
/ c
down and $30 a f '
monthly
month for 12 months. What is
19. Stock Dividend. A stock pays the compound interest rate
$4.00 every 6 months. What is monthly?
the present value of the stock if
(1 8)
Answers to Practice [ 1. 1000(1. 08) 2 + 1000(1.08)' + 1000 + lOOof- n ,^ 1 = $5823.50
DrnKlnmr L 0.08 J
Problems
9000
2 - P = =$97 ' 799 51
-
(1.045)^-l
Decisions and the Cost of Money
A printing company currently has the opportunity to purchase either of
two
printers to be operated for the next 5 years. The higher priced machine is
expected to have lower operating costs and thus will generate more profit,
and the lower priced machine may have to be overhauled after 3 years. If not
overhauled, it may break down and become unusable. The firm's cost of
money is 10 percent compounded yearly. The expected profit from each
machine is shown in the table. The rows indicate initial cost and profit under
various assumed conditions. Figures are in thousands of dollars.
Let's look at machine B under three circumstances:
1. No breakdown or overhaul
IS 16 15 12 II
+ 2 3 4 5
50 = $5.88
1.10 '
(I.IO) (1.10) (1.10) (1.10)
With overhaul
16 15 16
3 4 5
50 = $2.16
1.10 (1.10)- (I.IO) (1.10) (1.1 0)
With breakdown
IS 16 15 12
+ 2
- 50 -$0.95
1.10 (l.IO) (1-I0) 3 (1.10)-
It isseen that the present value of profit of the more expensive ma-
chine. $6.07. is greater than any other present value. Of course, an analysis
of circumstances involving risk requires the use of probability (to be studied

later in this course). Engineers must estimate the probability of an overhaul
or of a breakdown. However, this example should emphasize how the cost
of money enters into business decisions.
Summary and Review Exercises
This chapter contains mathematics commonly referred to as the "mathemat-

ics of finance." You should now be familiar with the following terms:
Important Simple interest (213) Amount of an annuity (231)

Principal (213) Sinking fund (234)
Terms
Compound interest (218) Amortization (241)
Compound period (218) Equation of value (245)
Compound amount (218) Perpetuity (247)
Effective rate (224) Deferred annuity (246)
Annuity (231) Geometric progression (227)
Present value of an annuity (238)
Important Formulas you have used in this chapter are as follows:
rormulas Simple interest and amount:
/ = Prt
A= P{\ + rt)
Compound amount:
A = P(l + if
Compound amount if a principal is compounded continuously:

A = Pe' r
Present value:
A
(1 + 0"
Amount of an annuity:
(1 + /)" ~ 1
A = R
i
Present value of an annuity:
I - (1 + Q-"
P = R
i
Present value of a perpetuity:
R
P = -
(1 + i) 1
The sum 5 of a geometric progression of n terms is
a(r" - 1)
Sn =
r - 1
where the first term is a and each term is obtained from the preceding terms
by multiplying by the constant r + 1.
Review Exercises 1. Find the simple interest due on a vestment of $4000 at 8% com-
loan of $5000 at 12% interest for pounded quarterly for 10 years.
3 months.
5. Find the effective rate equiva-
2. Find the compound interest and lent to 12% interest com-
compound amount for an invest- pounded semiannually.
ment of $4000 at 6% interest
6. Find the present value of $4000
compounded semiannually for
due in 5 years at 6% interest
10 years.
compounded semiannually.
3. Find the interest on $3000 at
7. Compute the present value and
12% simple interest for 6
the amount of an annuity of $400
months.
a year for 10 years at 8% interest
4. Compute the compound interest compounded annually.
and compound amount for an in-
8. Kalid obtained $190 from the $500 $500 $500

x
bank and signed a 3-month non- 1.08 (I.08) 2 (1.08)'
interest-bearing note for $200.
$500
Compute the simple interest rate
'
+
' '
12
(1.08)
he was charged.
15. A $10,000 debt is to be repaid by
9. Dr. Jones is planning to endow a
equal payments at the end of
chair at the university for
each 6 months for 3 years. If
$15,000 per year. Compute the
amount of
money is worth 8% inter-
the endowment if the
estcompounded semiannually,
fund is to be invested at 6% in-
make an amortization table for
terest compounded semiannu-
this debt.
ally.
16. A deposit of $500 is made to a

10. A contract pays $200 at the end bank at the end of each 6 months
of each quarter for 4 years and for 2\ years. If money is worth
$2000 additional at the end of the 8% compounded semi-
interest
last quarter. What is the present
make a table showing
annually,
value of the contract at 8% inter- how much money accumulates
est compounded quarterly? in the sinking fund.
11. What should be the semiannual 17. Find the compound amount the
deposit to a sinking fund estab- Native Americans would have if
lished to pay off a loan of $300 at they had invested $24 for 300
6% interest compounded annu- years at 8% interest com-
ally in 3 years if the fund pays pounded annually.
6% interest compounded semi-
annually?
18. A department store charges 2%
interest per month service
12. You wish to borrow $10,000 to- charge on unpaid balances. As-
day and $5000 5 years from now. sume that no payments are made
You plan to repay these loans for year and compute the ap-
1
with equal payments at the end proximate effective rate of inter-

of each year for 10 years. est.
If the
interest is 8% compounded an- 19. A house trailer was bought for
nually, what is your annual pay- $27,500, with $2500 paid down
ment? and the balance amortized at
12% interest compounded
13. Use geometric progressions to
monthly for 8 years. What is the
find the sum of the annuity
equity after the 50th payment?
$100 + $100(1.06) + $100(1.06) :
20. An orchard
will produce its first
+
+ $100(1.06) 8
crop end of 6 years. If af-
at the
14. Use a geometric progression to ter 6 years it is expected to pro-
find the present value of the an- duce an annual income of $6000
nuity
254 Chapter 4 Mathematics ot Finance
for 15 years, what is the cash 23. A company sets up a sinking

value of the orchard? Assume fund with yearly deposits of
thatmoney is worth 6% interest $10,000 to replace a piece of
compounded annually. equipment with a life of 8 years.
At an interest rate of 8% com-
21. Mark borrows $6000 today with
pounded yearly, how much
interest at 6% computed annu-
money will be available to re-
ally. He agrees to pay $1000 in 1
place the old equipment?
year, $2000 in 2 years, and the
balance 4 years from today. 24. Parents Paul and Joy are saving
Compute the final payment. money for Kate's college educa-
tion. How much must they save
22. You purchase an automobile for
yearly in order to have $10,000
$500 down and $300 a month for
when Kate enters college at 17
48 months. If interest14.2% is at
years of age if the money can be
compounded monthly, what is
invested at 8% compounded
the selling price of the car?
yearly?
Chapter Test
1. Find the present value of $1000 due in 3 years at 8% compounded

semiannually.
2. How much money will you have in the bank at the end of 3 years if you
deposit $500 at the end of each month for the 3 years? You draw 12%
interest compounded monthly.
3. If $1000 accumulates to $1200 in 6 months, what simple-interest rate is
being charged?
4. You borrow $2000 from the E-Z Loan Company. How much do you pay
back in 2 years if they charge 12% compounded monthly?
5. On a new automobile you agree to pay $100 every 6 months for 5 years.
Ifyou are charged interest at a rate of 8% compounded semiannually,
what is the cash value of the automobile?
6. You place your savings in a bank that pays a simple-interest rate of 5%.
If you have $2400 at the end of 4 years, how much did you deposit?
7. What is the effective rate of 12% compounded monthly?

8. You purchase a house for $120,000, pay $20,000 down, and amortize the
debt by monthly payments for 10 years. What are your monthly pay-
ments if money is worth 12% compounded monthly?
9. What is the present value of a perpetuity of $1000 every 6 months at 8%

compounded semiannually?
Chapter Test 255
10. You buy Your monthly payments are $1200. Make

a house for $100,000.
a schedule to show how you are reducing the loan each month for the
first 3 months if you pay 12% interest compounded monthly.
11. Paul secures a loan of $10,000. At the end of 1 year he repays $2000. At
the end of 2 years he pays $3000. At the end of 4 years he agrees to pay
payments made semiannually starting at the end
off the debt with 6 equal
of 4 years. bank charges 8% compounded semiannually, find the
If the
payments necessary to pay off the loan.
Counting Techniques and
Probability
Archaeological artifacts indicate that many of the early peoples played some
version of dice either for recreation or to determine the will of the tribal
deity. As more elaborate games were developed, the players began to ob-
serve certain patterns in the results, but they did not have the language of
probability to describe and analyze them. Over time, outstanding mathemati-
cians established probability as a legitimate field of inquiry, but the applica-
tion seemed to be for gamblers' games of chance.
As time passed, it became clear that probability was much more than
just a technique for gamblers. Physicists now use probability theory when
studying various gas and heat laws as well as in the theory of atomic physics.
Biologists apply the techniques of probability in genetics, the theory of
natural selection, and learning theory. Managers in government and industry
use probability techniques in decision-making and accounting processes.
Furthermore, probability is the theoretical basis of statistics, a discipline
that permeates most fields of study today.
In this chapter, you will come to understand the basic concepts of
probability. Counting techniques, such as the fundamental principle of
counting, tree diagrams, permutations and combinations, and additional set
terminology are used to assist you in computing probabilities.
256
The Language of Probability 257
^ 5.1 The Language of Probability
Overview One of the significant characteristics of our increasingly complex society is that we must deal
with questions for which there is no known answer but instead one or more probable (or
improbable) answers. Statements like "I am 90% confident that the mean weight is between 76.2
grams and 78.6 grams," "This cancer treatment has a 0.6 chance of leading to complete remis-
sion," and "The candidate has a 20.4% chance of winning the election" are common in conver-
sations today. A necessary skill for our times is the ability to measure the degree of uncertainty
in an undetermined situation. In this section, we discuss probability the language of uncer-
tainty. As the language of the undetermined and the uncertain, probability is an important tool
in many phases of our uncertain modern life. In this section we learn about:
Experiments
Outcomes
Sample space
Uniform sample space
Events
Properties of probability
Classical definition of probability
is a language of uncertainty, any discussion of proba-

Since probability
bilitypresupposes a process of observation in which the outcomes are not
predictable. Such a process is called an experiment. Any possible result of
an experiment is called an outcome.
EXAMPLE 1 Experiment: A coin is tossed.

Possible outcomes: Head (H) or tail (T)>
EXAMPLE 2 Experiment: A die (one member of a pair of dice) is tossed.

Possible outcomes: The top side of the die shows 1, 2, 3, 4, 5, or 6 dots.
Assuming is "fair," the outcomes (getting a head or

that the coin
getting a tail) inexperiment are said to be equally likely, because one
the first
outcome has the same chance of occurring as another. Likewise, the out-
comes in the second experiment (getting a 1, 2. 3, 4, 5, or 6 in the roll of a
die) are equally likely>
Before we can analyze an experiment, we must decide upon a sample

space (defined shortly), since different sample spaces can result from the
same experiment, depending on how the observer chooses to record the
outcomes.
EXAMPLE 3 Consider the experiment of drawing one card from six cards marked by
numbers through 6 and observing the number on the card. The set of
1
possible outcomes is
{1.2,3,4,5.6}
258 Chapter 5 Counting Techniques and Probability
EXAMPLE 4 In another experiment with the six cards, we might observe whether the
number is even or odd. Thus, the possible outcomes of this experiment are
{even, odd}
Notice that associated with the experiment of drawing a card are sev-
eral sets that classify the outcomes (Table 1).
TABLE 1
Set of Possible Classifications

Question of Outcomes
Is number on the card even or odd?

the {even, odd}
What is the number on the card? {1, 2, 3,4,5, 6}
Is the number on the card greater than
3? {less than or equal to 3, greater than 3}
Although the classifying sets are different, notice that they share cer-
tain properties. Ineach example, the set of outcomes classifies completely
or exhausts the possibilities of what can happen if the experiment is per-
formed. The number on the card is either even or odd; one of the numbers 1
2,3,4,5,6 must appear on the card; and the number on the card either is less
than or equal to 3 or is greater than 3. Moreover, the members of a set of
outcomes for an experiment are distinct; that is, they do not overlap.
Definition: Sample Space A sample space (denoted by S) is a set of outcomes of an experiment constructed in such
a way that
1. The categories do not overlap.
2. No result is classified more than once.

3. The set of outcomes is complete (exhausts all the possibilities).
EXAMPLE 5 A coin is flipped two times in succession.
Sample space A: One complete listing of the outcomes is
S = {(H,H), (H,T), (T,H), (T,T)}
The letter listed first in each pair indicates the result of the first flip, and the
letter listed second gives the result of the second flip.
Sample space B: An alternative way to list the outcomes is to ignore the

order in which the heads and tails occur and to record only how many of
each appear.
S = {(2H), (lHand IT), (2T)}

Another way to tabulate the same outcomes is to list the number of heads
that occur:
{0, 1,2}
The outcomes of this sample space are not equally likely. The chance that
heads occur is not the same as the chance that head occurs.|> 1
Now, for the outcomes of a sample space, we define a probability (or,
to be more precise, a probability function or rule). Probability is a rule that

assigns to each outcome in a sample space a real number between and 1,
inclusive.
Properties of Probability A probability rule on a sample space must satisfy two properties.
1. If A is an outcome, then the probability of A, denoted by P(A), is between and 1,
inclusive.
0<P(A)<1.
2. The sum of the probabilities of all outcomes in a sample space equals 1 or
P(S) = 1
These properties are illustrated in the examples that follow.
EXAMPLE 6 In the repeated tossing of a fair coin, one states intuitively that P(T) = \ and
/>(H) = i Notice that these probabilities satisfy the properties
(a) < P(T) < 1 and < P(H) < 1
(b) P(S) = />(H) + PCT) = i + 5=l
There is one whole class of sample spaces whose probability assign-

ments are particularly easy to determine.
Definition: Uniform If each outcome of the sample space is equally likely to occur, the sample space is called
Sample Space a uniform sample space.
Suppose a uniform sample space consists of m outcomes. Since each of

the outcomes is equally likely, it seems reasonable to assign to each outcome
A the same probability, denoted by P(A). Since the sum of the probabilities
of the m individual outcomes must be we note that 1 ,
P(A) + P(A) + P(A) + + P(A) = 1 or P(A) =
m times
Thus, each of the m outcomes has probability \lm.

Equal Probabilities In a uniform sample space with m outcomes, each outcome has probability Mm. This is
sometimes written
P(A)
n{S)
where n{S) represents the number of outcomes in a sample space.
EXAMPLE 7 Eight identical balls numbered 1 to 8 are placed in a box. Determine the
sample space and a probability rule describing the experiment of randomly
drawing a particular ball from the box.
SOLUTION A suitable sample space is {1, 2, 3, 4, 5, 6, 7, 8}, each number

representing one of the eight balls. each ball
If is equally likely to be drawn,
we assign a probability of g to each outcome.
P(\) = P(2) P&) = I
EXAMPLE 8 Notice in Figure 1(a) that the sample space consisting of {A, B, C, D} is a
uniform sample space. Thus, P(A) = P(B) = P(C) = P(D) = \. In Figure
1(b) the sample space is not a uniform sample space. Why not?
A
'
/
SOLUTION The sample space is tabulated as
{HH, HT, TH, TT}

The event of getting at least one head is {HH, HT, TH}. See Figure 2.
FIGURE 2
EXAMPLE 12 Consider the experiment of tossing two coins. What is the probability of
tossing two heads? At this juncture, we must be careful. We have discussed
two sample spaces for this experiment:
{HH, HT, TT} or {2H, 1H, OH}
in which we do not observe which coin has a head and which coin has a tail.
In the second sample space
{HH, HT, TH, TT}
the ordered pair indicates which coin has a head and which coin has a tail.
Note that we cannot apply the definition of probability on the sample space
in the first case, because it is not a uniform sample space; however, the
definition is applicable to the second sample space. Now let A be the event of
tossing two heads. Using the second sample space,
Pi A) = n(A)
n(S)
Practice Problems D> 1. Write a set of outcomes for an experiment associated with the following
spinner.
A sack contains 5 chocolate candies, 3 butterscotch candies, and 1 pep-

permint candy. One candy is drawn from the sack and eaten. Then a
second candy is drawn and eaten. What is a sample space for the experi-
ment?
Consider again the experiment of drawing a card from a set of six cards
numbered through 6.
1
(a) List three events associated with this experiment that involve only
one element.
(b) List three events that involve more than one element.
Consider the two following spinners. If both spinners are activated, find
the probability of spinning a total of 5 points.
Exercise Set 5.1 1. Give the sample space for each

of the four spinners.
Assign a probability to each out- Sally,Tom, Ben, Nakesha, and

come in the sample spaces of Louis place their names in a hat.
Exercise 1. One name is to be drawn to de-
termine who will wash the auto-
A Find a sam-
fair die is rolled.
mobile. Tabulate a sample space
ple space for this experiment
and a probability rule.
and assign a probability to each
outcome.
5. What is the probability of getting which answer is correct, so you

heads when a two-headed coin is randomly choose one of the 5
tossed? answers. What is the probability
thatyou select the correct an-
6. In Exercise 5, what is the proba-
swer? An incorrect answer?
bility of getting tails?
Suppose that there is an equally 14. Which of the following could not
likely probability that the spinner be a probability? Why?
will stop at any one of the 6 sections (a) -\
for the given spinner. (b)
(c) 0.001
(d)
(e) 1.03
(f) 0.01
(g) I
(h) 1
15. List the elements in a sample

space for the simultaneous toss-
ing of a coin and drawing of a
7. What is the probability of stop- card from a set of six numbered
ping on an even number? 1 through 6.
8. What is the probability of stop- 16. A box contains 3 red balls and 4
ping on a multiple of 3? black balls. The balls are not dis-
tinct from one another. Let R
9. What is the probability of stop-
represent a red ball and B a
ping on an even number or a
black ball. Tabulate a sample
multiple of 3?
space if
10. What is the probability of stop- (a) One ball is drawn at a time.
ping on an even number or an (b) Two balls are drawn at a
odd number? time.
(c) Three balls are drawn at a
11. What is the probability of stop-
time.
ping on a number less than 3?
(d) Are these sample spaces uni-
12. A card is drawn from an ordi- form?
nary deck. What is the probabil-
17. Three coins are tossed, and the
ity of getting
number of heads is recorded.
(a) A heart?
Which of the following sets is a
(b) An ace?
sample space for this experi-
(c) The jack of spades?
ment? Why do the other sets fail
(d) A red card?
to qualify as sample spaces?
(e) A red ace?
(a) {1,2,3}
13. A multiple-choice question has 5 (b) {0, 1,2}
possible answers. You haven't (c) {0, 1, 2. 3,4}
studied and hence have no idea (d) {0, 1,2,3}
18. A box contains four balls num- 20. One card is drawn from each of
bered 1 to 4. Record a sample two sets of four cards (each set

space for the following experi- numbered through 4). List a
1
ments. uniform sample space to show

(a) A ball is drawn and the num- that the two cards can be drawn
ber is recorded. The ball is in 16 different ways. What is the
returned and a second ball is probability that the sum of the

drawn and recorded. numbers is
(b) A ball is drawn and re- (a) Greater than 4?

corded. Without replacing (b) Equal to 9?
the first ball, a second ball is (c) Equal to 7?
drawn and recorded. (d) Greater than 7?
19. A box contains 4 black balls, 7 21. Two coins are tossed. What is
white balls, and 3 red balls. If a the probability of getting

ball is drawn, what is the proba- (a) Two heads?
bility of getting the following (b) Exactly one head?
colors? (c) At most one head?
(a) Black
22. In drawing a card from a stan-
(b) Red
dard deck of cards, you reason
(c) White
that you might get a spade (S) or
(d) Red or white
might not get a spade. There-
(e) Black or white
fore, there are two outcomes.
(f) Red or white or black
Hence, P(S) = . Is this reason-
ing correct?
Applications
(Business and 23. Quality Control. A shipment is 24. Executive Boards. An executive
Economics) believed to contain 100 good ar- board of a corporation is made
ticles, 5 articles with minor de- up of 5 members whom we shall
fects, and 3 articles with major call A, B, C, D, and E. A com-
defects. If 1 article is drawn mittee of 3 is chosen to select a

from the shipment, what is the president for the corporation.
probability that it will Find outcomes of the sample
(a) Not have a defect? space that represent all the pos-
(b) Have a defect? sible committees.
(c) Have a major defect?
and Life Sciences) 25. Drug Analysis. A medical re- drugs, V, W, X, Y, and Z, for
search institute is experimenting concentrated research.
with possible cures for cancer. (a) List the sample outcomes
Dr. Stewart, the scientist in where Z is 1 of the chosen
charge of the experiment, ini- drugs.
tially selects 3 of 5 possible (b) List the sample outcomes
where X and Y are both (d) Dr. Stewart decides that a

among the chosen drugs. new drug, drug A, will defi-
(c) Another scientist suddenly nitely be 1 of the 3 drugs
announces that drug V is used for experimentation,
definitely not a cure for can- She now needs to choose 2
cer. If drug Vis dropped, list of the 4 possible drugs. List
the sample space from which the sample outcomes where
Dr. Stewart can choose 3 of W is one of these chosen
4 possible drugs. drugs.
Answers to Practice 1. The set of possible outcomes of spinning and then observing the color
where the needle stops is {red, black, white, green}.
2. S = {(C, C), (C, B), (C, P), (B, C), (B, B), (B, P), (P, C), (P, B)}.
3. (a) Some possible events involving one element are: observing a 1; ob-
serving a 2; observing a 3; and so on.
(b) Events containing more than one element are: observing a number
observing an odd number; observing a number divisible
less than 4;
by 2; and so on.
4. S = {(1, 1), (1, 2), (1, 3), (1,4), (1, 5), (1, 6), (2, 1) . . . (2, 6) (3, 1) . . .
(3, 6) (4, 1) . . . (4, 6)}
n(S) = 4
6 = 24
E= = 4
.
{(1,4), (2, 3), (3,2), (4,
P(5 points)
,
= n{E)
= -=-
4 1
1)} n(E)
5.2 Empirical Probability and the Fundamental Principle

of Counting
A poor way of assigning probabilities to outcomes in a sample space is subjective judgment; yet
subjective judgment is often the only tool accessible in certain situations. Hence a sales man-
ager might project that "We have a probability of j of getting the XYZ contract," and the
ubiquitous man on the street might prophesy that "The probability of Dan Quick's being elected
is m" Generally, such assignments are merely measures of the strength of a person's belief
from previous experience. If this experience actually uses a relative-frequency concept, the
number assigned for a probability could be close to the actual probability.
In this section, we shall study a rule for assigning probabilities based on empirical data.
Then we will consider counting procedures that assist in assigning probabilities. We need to
understand:
Relative frequency
Empirical probability
Tree diagrams
The fundamental principle of counting
Empirical Probability and the Fundamental Principle of Counting 267
In the previous section we assigned probabilities to outcomes using a

definition involving the number of ways an event could occur. In this sec-
tion, we will assign probabilities based on what we call empirical data. To
introduce this concept consider the following example.
EXAMPLE 13 A fair die is rolled 10,000 times. Table 2 itemizes the number of times a 1 has
occurred at various stages of the process. Notice that as TV becomes larger,
the relative frequency stabilizes in the neighborhood of 0.166 = |. Thus, we
are willing to assign the probability
/>(!) =
|
Using the sample-space definition of probability, we note that 5 {1,2,
3.4, 5,6}and = {1}. Thus,
n(E) _ 1
P(\) ~
n(S) 6
which is the same answer as obtained empirically.
TABLE 2
Definition: Empirical
Probability
EXAMPLE 14 A loaded die (one which outcomes are not equally likely) is thrown 7000
in
times with the results shown in Table 3. Determine a rule for assigning a
probability to each outcome.
TABLE 3
Counting Procedures
We now consider procedures that assist in finding probabilities.
EXAMPLE 16 The college chorale is planning a concert tour with performances in Dallas,
St. Louis, and New Orleans. In how many ways can they arrange their
itinerary?
SOLUTION If there is no restriction on the order of the performances, any 1 of

the 3 cities can be chosen as the first stop. After the first city is selected,
either of the other cities can be second, and the remaining city can be the last
stop. A tree diagram can aid the chorale in determining their possible tour
schedule. Starting at the campus, draw lines to each of the 3 cities. Then
draw lines from these cities to each of the remaining choices. The result is a
diagram resembling the branches of a tree (see Figure 3).
FIGURE 3
Whenever a task can be done in two or more stages and each stage can
be done in a number of ways, a tree diagram provides a good illustration of
the choices involved and serves as an aid in determining the number of ways
the whole task can be accomplished.
Now let's consider again the chorale in Example 16.
EXAMPLE 17 The members of the chorale decided to sing first in New Orleans, next in
Dallas, and finally in St. Louis. Now they must decide on their modes of
transportation. They can from the campus to New Orleans by bus or
travel
plane; from New Orleans to Dallas by bus, plane, or train; and from Dallas to
St. Louis by bus or train. The tree diagram in Figure 4 indicates the different
modes of transportation. The first part of the trip can be made in 2 ways, the
second part in 3 ways, and the last part in 2 ways. Notice that the number of
ways of transportation that can be chosen is 2 3 2 = 12 ways.
FIGURE 4
This example introduces the following principle.
The Fundamental 1, If two experiments are performed in order with n, possible outcomes of the first
Principle of Counting experiment and n 2 possible outcomes from the second experiment, then there are
n-i -n 2
combined outcomes of the first experiment followed by the second.
2. In general, if k experiments are performed in order with possible number of outcomes

n u n 2 ,n 3 nk , respectively, then there are
n,n 2 n 3 -
. . . -n k
possible outcomes of the experiments performed in order.
The fundamental principle of counting is helpful in solving problems

such as the following.
EXAMPLE 18 In many automobile license plates have on them a combination of 3

states,
letters digits. If all letters and digits may be used repeatedly,
followed by 3
how many combinations are available to each of these states?
SOLUTION There are 26 letters to choose from for each of the 3 letter places,
and there are 10 digits to choose from for the digit places. By the fundamen-
tal principle of counting, the number of combinations is
26
26
26
10 10
10 = 17,576,000
EXAMPLE 19 If a couple plans to have 3 children, what is the probability that exactly 2 will
be boys? (Assume that it is equally likely for a child to be a boy or a girl.)
SOLUTION The tree diagram in Figure 5 indicates 8 paths, or ways that the
couple can have 3 children. Each of these is equally likely. So the sample
space consists of 8 equally likely outcomes. By counting, you note that 3 of
these have exactly 2 boys, or the event can happen in 3 ways. Therefore,
/êxactly 2 boys) =
FIGURE 5
EXAMPLE 20 A card is drawn from deck of 52 cards. Then the card is replaced, the deck
a
is reshuffled, and a second card is drawn. What is the probability of an ace
on the first draw and a king on the second?
SOLUTION Since the is returned, there are 52

first card 52 ways of drawing
the 2 cards, so n( S) = 52
There are 4 ways of drawing an ace on the first

52.
draw and 4 ways of drawing a king on the second draw, or 4 4 ways of
drawing both, or /?() = 4-4. Thus,
n(E)
/"(ace followed by a king)
(a) What is the probability that an employee selected at random is in
group C?
(b) What is the probability that an employee selected at random has no
opinion?
(c) What is the probability that an employee selected at random is in
A?
group
2. How many different color combinations can you have for a two-toned car
(top one color and body another color) using red, white, and black paint?
(Use a tree diagram.)
3. A coin is tossed twice. Draw a tree diagram to illustrate the possible
outcomes.
4. A box contains and 7 white balls. A ball is drawn, its color is
5 red balls
noted, and a second ball drawn. How many ways can 2 balls be drawn?
is
What is the probability of drawing a red ball followed by a white ball?

Note: The first ball is not returned to the box.
Exercise Set 5.2 l. There are 6 roads from A to B how many ways could you pur-
and 4 roads between B and C. chase a meal?
(a) In how many ways can Joy
4. Hue and May play in a Ping-
drive from A to C by way of
Pong tournament. The first
B? player to win 4 games wins the
(b) In how many ways can Joy tournament. Make a tree dia-
drive round trip from A to B gram showing the possible ways
to C and return to A through in which the tournament can
B? turn out.
2. Kate wants to buy an automo-
5. A die is tossed and a chip is
bile. She has a choice of 2 body
drawn from a box containing 3
styles (standard and sports
chips numbered 1, 2, and 3.
model) and 4 colors (green, red,
How many possible outcomes
black, blue). In how many ways
can be obtained from this experi-
can she select the automobile?
ment? Verify your answer with a
3. A restaurant offers the following tree diagram.
menu.
6. By means of a tree diagram, ana-
Beverage
lyze the number of possibilities
Main Course Vegetables
in tossing a coin 4 times.
Beef Potatoes Milk
Ham Green beans Coffee 7. In how many ways can 2 speak-
Fried chicken Green peas Tea ers be arranged on a program?
Shrimp Asparagus
8. In how many ways can 3 speak-
main course, 2 ers be arranged on a program?

If you choose 1
vegetables, and 1 beverage, in

9. In how many ways can 4 speak- (b) Adjacent letters cannot be

ers be arranged on a program? alike?
(c) Letters can be repeated?
10. A box contains 6 different-col-
ored balls: red, white, blue, 14. Two cards are drawn from a
black, green, and yellow. If 2 standard deck of playing cards.
balls are drawn at random, one What is the probability that a
at a time, and replaced, what is king is drawn followed by an ace
the probability of getting (a) If the first card is replaced
(a) A yellow ball followed by a before the second is drawn?
red ball? (b) If the first card is not re-
(b) A red ball followed by a blue placed before the second is
ball? drawn?
11. Four coins are tossed. What is
15. Exam. Ousley Heating and
the probability of getting
Cooling has established the fol-
(a) Four tails?
lowing distribution of monthly
(b) Exactly 2 tails?
service calls over the last four
(c) At least 3 heads?
years.
(d) Exactly 1 head?
12. An ice chest contains 5 cans of

cola, 7 cans of ginger ale, and 3
cans of root beer. Al randomly
selects a can, and then Sheila
takes one. What is the probabil-
ity that
(a) Al gets a cola and Sheila a

root beer?
(b) Al gets a ginger ale and
Sheila a cola?
(c) Both get root beers?
(d) Neither gets a cola?
Hint: There are 15 14 ways the
experiment can occur.
13. How many 4-letter words are

possible out of the alphabet if
(a) No can be used more

letter
than once?
the top 25% in sales last year,

the next 25%, and the lowest
50%.
and Life Sciences)
22. Politics. Suppose that in a local ee's age and the number of on-
election only 2 parties are repre- the-job accidents. The following
sented, D and R. Draw a tree di- table summarizes the findings.
agram illustrating 4 consecutive
elections and determine how Number ot Accidents
many possibilities result in Age Group 3 or More
(a) At least 1 party change. Under 20 18 22 8 12

(b) No party changes. 20-39 26 18 8 10
(c) Exactly 1 party change. 40-59 34 14 8 6
(d) More than 2 party changes. 60 and over 42 10 12 2
23. Medicine. Paul's clinic indi-

(a) What is the probability that
cates a cure has been found for a
an employee is in the 20-39
certain blood disease. Out of age group?
80,000 patients, 74,000 recov-
(b) What is the probability that
ered after using the medication.
an employee will have 2 ac-
Assign a probability that a per-
cidents?
son suffering from the blood dis-
(c) What is the probability that
ease will recover using this med-
an employee will have more
icine.
than 2 accidents?
24. On-the-job- Accidents. A sociol- (d) What is the probability of an
ogy class made a study of the employee having an acci-
relationship between an employ- dent?
10''
ctice
1. (a) /'(employee is in group C) = j
5U "
Problems
(b) /'(employee has no opinion) = -^~
. . 210
(c) ftemployee is in group A) = -r-
2. Six ways 3.
There are 12
11 = 132 ways of drawing two balls so n(S) = 132. There
are 5
7 = 35 ways of drawing a red ball and then a white ball.
35
/"(red ball followed by a white ball) = -prx
5.3 Counting Techniques Using Permutations

and Combinations
Overview The fundamental principle of counting can be used to derive two extremely important special
cases for counting techniques: permutations and combinations. Both of these concepts are
useful not only in solving complicated probability problems but also in other types of applica-
tion problems. The main things to remember are that the order of arrangements is important for
permutations and that combinations do not involve the order of arrangements.
Suppose we have four letters,
{A, B, C, D}
and list them two at a time:
{AB, AC, AD, BA. BC, BD, CA, CB, CD, DA, DB, DC}
Note that AB is listed and so is BA. Likewise, BD and DB are listed. In this
listing the order is important. When order is important, arrangements of
objects are called permutations.
Definition: Permutations The ordered arrangements of r objects selected from a set of n different objects (r ^ n)
are called permutations.
EXAMPLE 21 Tabulate the permutations of {A, B, C} taken three at a time.
SOLUTION There are six different permutations:
ABC BAC CBA

ACB BCA CAB
EXAMPLE 22 How many permutations are there of the set {A, B, C, D}?
SOLUTION The permutations of four objects implies that they are taken 4 at a
time. First we draw a tree diagram (Figure 6).
CDBDBCCDADACBDADABBCACBA
DCDBCBDCDACADBDABACBCAAB
FIGURE 6
Counting Techniques Using Permutations and Combinations 277
The can be chosen in four ways. After this occurs, the second
first letter
letter can be selected in three, or the first two letters can be selected in 4 3
ways. Then the third letter can be selected in two ways, or the first three
letters can be selected in 4 3 2 ways. Finally, the fourth letter can be

selected in only one way, so there are 4 3 2 1 permutations of four
letters.
In general.
Number of Permutations The number of permutations of n distinct objects is
of n Objects
n(n - 1)(n - 2) . .
. -3-2 1
which can be written as n\ (read n-factorial).
The preceding discussion suggests n\ to be the product of positive

integers to n, inclusive. The product 6 5 4 3 2
1
1 may be denoted by 6!,
called 6-factorial. We define both 1 ! and 0! to be 1 . The statement that 0! = 1

may seem surprising, but you will learn later in your work with factorials
that this definition is reasonable and consistent with the factorial idea for
positive integers.
Definition: Factorial n\ is called n factorial and is defined as
n! ='n(n - 1)(n - 2) 3-2-1
for n a positive integer. Also, 0! = 1.
CALCULATOR NOTE Please note that on most calculators, there is a key [Tj. [vT], or [n\]. Use this
key to verify that 8! is 40.320 and 10! = 3,628,800.
EXAMPLE 23 Six workers are assigned 6 different jobs. In how many ways can the assign-
ments be made?
SOLUTION 6! = 720 ways

Our objective now is to develop a formula for the number of permuta-
tions of n objects taken r at a time, denoted by P(n, r). Consider the follow-
ing example.
EXAMPLE 24 A first-year class is to elect a president, a vice-president, a secretary, and a

treasurer from 6 class members who qualify. In how many ways can the
class officers be selected?
solution If we consider the order as president, vice-president, secretary,

and treasurer, then {Maria, Tom, Jim, Tomoko} is certainly different from
{Tomoko, Tom, Maria, Jim}. Thus, the answer to the problem is P(6,4), the
number of permutations of 6 things taken 4 at a time. Use the fundamental
principle of counting and note that the position of president can be filled in 6
ways. After this occurs, the position filled in 5 ways,
of vice-president can be
or the 2 positions can be ways. Then the secretary can be
filled in 6
5
selected in 4 ways, or the 3 positions in 6 5 4 ways. Finally, there are only
3 people left to be selected for treasurer. Hence, the number of ways that all
4 positions can be filled is 6 5 4 3, or
P(6A) = 6
5
4
3
To express P(6,4) in terms of factorials, multiply and divide by 2!
6
5
4
3
2!
P(6,4) =
2!
Now we can write

6! 6!
P(6A) as r-r
2! (6 - 4)!
in order to express everything in terms of the only numbers (6 and 4) given in
the example.
By reasoning in the same way, we find the following:
Number of Permutations
CALCULATOR NOTE Some books use !', instead of P(n,r). Also, some calculators have a \P r \
key. Determine /> (4.2) with your calculator.
4CZD2H
Objects Not Distinguishable
Suppose we wish to find the number of arrangements of n objects where not

all objects are distinguishable. For example, suppose we permute the letters
in seem.
seem
Number of Permutations If we have a set of n objects to be arranged when there are n, of one indistinguishable
(Indistinguishable type, n 2 of a second indistinguishable type, continuing until there are n k of the /rth
Objects) indistinguishable type, then the number of possible permutations of the n objects is
ni! n2 \ nk \
where
n-\ + n2 + + nk = n.
Using this rule, the number of arrangements of the letters of seem is
4!
12
2! 1! 1!
as shown previously.
Combinations
Let's turn our attention now to the arrangement of elements without regard
to order. Let's list four letters {A, B, C, D} two at a time without regard to
order
{AB, AC, AD, BC, BD. CD}
These selections of four letters taken two at a time are called combinations
and denoted by C(4,2).
EXAMPLE 25 The first-year class is to elect 4 class officers from 6 class members who
qualify. How many sets of class officers are possible?
solution The answer is of course C(6,4), the number of combinations of 6

things taken 4 at a time. The order does not matter. In this case {Mai,
Toshio, Ron, Jill} is the same as {Jill, Toshio, Mai, Ron}. By writing down
the different sets of 4 officers from 6 prospects, let's say, {M, T, J, R, B, C},
we get the following possibilities:
{M, T, J, R}, {M, T, J, B}. {M, T, J, C}, {M, T, R, B}, {M, T. R, C},
{M, T, B, C}, {M, J, R, B}, {M, J, R, C}, {M, J, B, C}, {M, R, B, C},
{T, J, R, B}, {T, J, R, C}, {T, J, B. C}, {T, R, B, C}, {J, R, B, C}
We see that
C(6.4) = 15
Let's obtain the answer another way. Take each combination of offi-
cers and arrange the 4 officers as president, vice-president, secretary, and

treasurer. There would be /*(4,4) arrangements. Now if we multiply this

result by the number of selections of 4 officers, C(6,4), the answer should be
the number of arrangements of 6 things taken 4 at a time, or
P(6,4) = C(6,4)
P(4,4)
Thus,
C(6,4) = P(6A)
There are 13 hearts. The number of ways in which 5 cards can be drawn from
these 13 is C(13,5) = 1287.
1287
P(a\\ hearts) = - .0005
2,598,960
EXAMPLE 27 A quality-control inspector at the GE factory randomly selects 5 bulbs from

each lot of 100 bulbs that is produced and inspects them for defects. If a lot
has 96 good bulbs and 4 defective bulbs, what is the probability that the
inspector will find 3 good bulbs and 2 defective bulbs in her sample?
SOLUTION The sample space consists of all possible combinations of 5 bulbs

that can be chosen from 100. Hence, the number of outcomes in the sample
space is C( 100,5). One outcome is the event where exactly 3 of the bulbs are
good and 2 are defective. The 3 bulbs are selected without regard to order
from 96 good bulbs in C(96,3) ways. In like manner, the 2 defective bulbs are
selected from 4 in C(4,2) ways. By the fundamental principle of counting,
there are C(96,3) C(4,2) outcomes in the event. Thus,

C(96,3)C(4,2)
P(E) 0.01
C( 100,5)
Practice Problems t> 1 . In how many ways can a president, a vice-president, and a secretary be
selected from 10 people?
2. Find the number of arrangements of the letters in toot.
3. A box contains 4 red balls and 6 other balls. Two balls are drawn. What is
the probability that they are red?
Exercise Set 5.3 l. Consider the 4 objects {W, (e) P(6,6)

X, Y, Z}. (f) IV, 5)
(a) How many combinations of (g) C(7,5)
two objects can be chosen (h) C( 100,89)
from this set? List them. (i) C(100,ll)
(b) How many combinations of
3. Evaluate each of the following:
3 objects can be chosen from
(a) C(10,6)
this set? Listthem.
(b) C(10,0)
(c) List all combinations of 1 ob-
(c) C(15,l)
ject chosen from this set.
(d) C(4,2)
(e) C(r,2)
Write in factorial notation.
(f) C(rj - 1)
(a) P(5,3)
(b) C(4,2) 4. Evaluate each of the following:
(c) C(44,22) (a) P(S,3)
(d) C(8,8) (b) P(6,5)
(c) />(8,1) by 3 numbers (examples: MFT-

(d) P(9,2) 986, APT-098). How many dif-
(e) P(7,2) ferent license plates can be man-

(f) />(8,7) ufactured if no repetitions are
allowed?
5. In how many ways can 5 speak-
ers be arranged on a program? 13. Employee I.D. numbers at a
large factory consist of 4-digit
6. A coach of a football team must
numbers such as 0133, 4499, and
choose a first-string quarterback
0000.
and a second-string quarterback
(a) How many possible I.D.
from 8 aspiring superstars. In
numbers are there?
how many ways can the choice
(b) How many possible I.D.
be made?
numbers are there in which
7. In how many ways can 7 stu- all 4 digits are different?
dents line up outside Professor
14. A bowl contains 8 red marbles
Smith's door to complain about
and 14 black marbles. Three
grades?
marbles are selected at random
8. A student-body president is from the bowl without replace-
asked to appoint a committee ment.
consisting of 5 boys and 3 girls. (a) What is the probability that
A list of 12 boys and 10 girls is all 3 are black?
provided from which to make (b) What is the probability that 1
the appointments. How many is red and 2 are black?

different committees can be se- 15. A hat contains 20 slips of paper
lected? numbered 1 to 20. If 3 are drawn
9. How many hands con-
different without replacement, what is the
sisting of 7 cards can be drawn probability that all are numbered

from an ordinary deck of cards? less than 10?
16. The license plates of a certain

10. A special committee of 3 per-
state display 3 letters followed
sons must be selected from a 12-
by 3 numbers.
person board of directors. In
how many ways can the commit-
the letters of a randomly se-
tee be selected?
lected tag read HOT?
11. Write a simple expression for (b) What is the probability that
each of the following: none of the digits on the tag
(a) P(r,l) exceeds 4?
(b) P(k,2)
17. Employee I.D. numbers at a
(c) P(r,r - 1)
large factory consist of 4-digit
(d) P(k,k - 2)
numbers. What is the probability
(e) P(k,3)
- that if a number is chosen at ran-
(f) P(k,k 3)
dom from the list of I.D. num-
12. The license plates for a certain bers, all 4 of its digits would be
state display 3 letters followed different?
18. A typical social-security number that a given variety is repre-

is 413-22-9802. If a social-secu- sented'.'
rity number is chosen at ran-
12 11
dom, what is the probability that (a) ,d
,
all the digits will be the same? O20 >

46
19. From a standard deck of cards, (b) (e)

24'
how many different hands con-
sisting of 7 cards can be drawn
consisting of (c)
C(24,12)
(a) Seven spades?
(b) Five clubs, 2 hearts? 22. Exam. A random sample of 6 is
(c) Four clubs, 1 spade, 2 selected with replacement from

hearts? an urn that contains 10 red. 5
(d) Three clubs, 2 hearts, 2 dia- white, and 5 blue balls. What is
monds? the probability that the sample

contains 2 balls of each color?
20. Exam. An urn contains 10 red,
20 white, and 30 blue balls. If 6
(a)
balls are drawn at random with- 1024
out replacement, what is the
probability that 1 ball is red. 2
are white, and 3 are blue?
la) (d)
36 18
23,
549
(e)
649
21 Exam. A box contains 12 vari-

eties of candy and exactly 2
pieces of each variety. If 12
pieces of candy are selected at
random, what is the probability

ders are submitted at one time. (d) What is the probability that
In how many ways can 2 orders no defective tires will be
be submitted to a local firm and 2 among the 3 selected?
to an out-of-state firm?
27. Organizational Management.
26. Quality Control. In a quality- Among the 30 employees in the
control check at the Acme Tire Espy Corporation, there are 20
Company, 3 tires are randomly women and 10 men; there are 18
selected and inspected from employees, and
pollsters, 6 sales
each lot of 20 tires produced. 6 management employees.
Suppose a lot contains 4 defec- Three employees are chosen to
tive tires and 16 good tires. form a committee. In how many
(a) In how many ways can 3 ways can a committee be se-
tires be selected from 20? lected to have
(b) How many different selec- (a) Two women and man? 1
tions of 3 tires from 20 will (b) Two pollsters and 1 sales-
contain exactly 1 defective person?

tire? (c) One salesperson, 1 pollster,
(c) What is the probability that and 1 management em-

exactly 1 of the 3 tires se- ployee?
lected will be defective?
and Life Sciences) 28. Dispensing Drugs. The order of be performed using different
administering 5 different drugs is groups of 3? (A rat may be in
important. different groups of 3 rats.)
(a) In how many ways can all 5
30. Medicine. Assume there are 8
drugs be administered?
classifications of blood types. In
(b) In how many ways can 3 of
samples of 5, how many possible
the 5 drugs be administered?
samples exist so that all classifi-
29. Experimental Design. Ten rats cations are different and no 2
are selected for an experiment. samples have exactly the same 5
Each trial run is to involve 3 rats classifications?
at a time. How many trials can
Answers to Practice 10!

Problems
,

,.,
W,3) = ^= 720
2!2!
C(4,2)
3. P(2 red) = =
"
2_
C(10,2) 15
Probability for the Union of Events 285
5.4 Probability for the Union of Events
Overview Consider the following problems. A card is drawn from a standard deck of cards. What is the
probability that it is either an ace or a spade? Of the first-year students who entered Samson
University last year, 12% failed first-year English, 16% failed mathematics, and 6% failed both
English and mathematics. An admissions counselor would like to know what percentage failed
English or mathematics. In both of the preceding examples the first event has parts in common
with the second event. In the first example, the ace of spades is common. In the second
example, those who fail both mathematics and English are common. The new notations intro-
duced in this section will help us find solutions to problems like these. We will study
Intersection and union of sets

Probability of /A or 8
Probability of A and B
Probability that A does not occur

Odds
As an introduction to the study of the union of events, we consider a

review of special notations that are used for discussing the relationship
among members of two or more sets.
Definitions: Intersection 1. The intersection of two sets A and 6 (denoted by A n 8) is the set of all elements
and Union of Sets common to both A and B.
2. A and B are any two sets,
If the union of A and 6, denoted by A u S, is the set
consisting of all the elements in set A or in set S or in both A and B.
EXAMPLE 28 Let A represent a committee of {Joe, Dave, Sue, John, Jack} and B represent
a second committee of {Sue, Edward, Cecil, John}. The intersection of these
two sets, A n B is {Sue, John}. The union of these two sets is
A U B = {Joe, Dave, Sue, John, Jack, Edward, Cecil}
Recall that A U B is the set of all elements that belong to A or B, or to

both A and/?. If there are elements common to both sets, they are listed only
once in the union. Given
A = {a, b, c, d, e} and B = {c, d, e,f, g}
then
Al) B = {a,b, c, d, e,f, No element listed twice
The shaded regions in Figure 7 compare intersection and union under

different situations for sets A and B. Note in part (a) that A and B overlap or
have elements in common. In (b), A is a subset of B. In (c), A and B have no
elements in common, or A n B = 0.
FIGURE 7
We are now ready to consider some additional configurations of

events. Three of these relationships are of such importance that we list them
as special events.
Definitions: And, Or, and 1 The event A u B (A or 8) is the collection of all outcomes that are in A or in B or in both
Complement A and B.
2. The event Af~\B(A and 6) is the set of all outcomes that are in both A and B.
3. The complement of an event A, denoted A', is the collection of all outcomes that are in
the sample space and are nor in A.
We illustrate these concepts with examples involving the roll of a die.
EXAMPLE 29 In the rolling of a fair die, what is the probability of getting either an odd
number or a 4?
SOLUTION We let O represent an odd number and F represent a 4 and seek

P(0 U F). In Figure 8(b) we see that
n(0 U F) 4 2
P(0 U F) =
n(S) 6 3
F/GUflE 8
Note in Figure 8(a) that
P(O) = \ and P(F) = \
Thus,
/>(0 U F) = P(O) + PtF)
4
= 3 1
" +
6 6 6 >
EXAMPLE 30 In the rolling of the same fair die, what is the probability of getting either an
even number or a 4?
solution Let E represent an even number and F represent a 4. We seek

P(E U F). In Figure 9(b) we see that
n(E U F) _ 3
P(E U F) =
n(S) ~ 6
(b)uF
F/GUfl 9
Note in Figure 9(a) that
P(E) and P(F) = 7

6
P(E\JF) + P(E) + P(F)
6 6
What is the difference between the problems in the two previous exam-
ples? For P(0 U F), F and O had no points in common. For P(E UF), and
F overlapped. This discussion suggests the following definition and property
of probability:
Mutually Exclusive Events 1. Events A and B are mutually exclusive if they have no outcomes in common.
2. If events A and B are mutually exclusive, P(A U 8) = P(A) + P(B)
EXAMPLE 31 From a standard deck of cards, we draw one card. What is the probability of
getting a spade or a red card?
SOLUTION Note that these events are mutually exclusive.
P(SP)
13
52 =
26
52
P(SP U R) = P(SP) + P(R)

13 26 3
=
~ + =
"
52 52 4
Now let's return to the preceding example, where we noted that P(E U
F) = f, P(E) = h and P(F) = I Why is P{E U F) = P() + P(F)? These
events are not mutually exclusive. The outcome 4 is in both E and F, and is
thus counted twice in P(E) + P(F) (see Figure 9(a)). The probability that 4 is
in both E and F is
P(E fl F) =
Since D F is included twice in P(E) + F(F), we subtract one of these and

note that
P(E U F) = F() + F(F) - F( n F)

3 1
+
6 6
We can generalize this concept by realizing that in set theory the num-
ber of outcomes in event A or in event B is the number in A plus the number
in B less the number in A n B, which has been counted in both A and B (see
Figure 10). Thus,
n(A U B) = fl(A) + n{B) - n(A D B)
FIGURE 10
Divide both sides of the equation by N, the number of elements in a sample

space, to obtain
n{A U B) _ n(A) n{B) n(A n B)

N N N N
or
P(A UB) = P(A) + P(B) - P(A n B)
Probability or A or B For any two events A and 8, the probability of A or fi is given by
P{A(JB) = P(A) + P(B)-P(Af)B)
Note if A and B are mutually exclusive, P(A n B) = 0, giving the

formula on page 288.
EXAMPLE 32 Of the freshmen at Hard College last year, 12% failed English, 8% failed
history, and 4% failed both English and history. What percent failed English
or history?
SOLUTION P(E) = .12
P(H) = .08
P{E n H) = .04
P(E \JH) = P(E) + P(H) - P(E n H)

= .12 + .08 - .04
= .16
Sixteen percent failed English or history>
The preceding discussion can be extended to 3 events, A, B, and C.

Probability of A or B or C P(A U B U C) = P(A) + P(B) + P(C) - P(A fl B)
-P(y\nc)-P(Bnc) + P(/\nBnc)
In Figure
1 1 note that A B, and C contain the shaded region 3 times (in
, ,
A, and in C), and -(A n C) subtracts one of these regions. So do

in B,
-(A n B) and -(B D C). Thus, this region has been deleted. In order to
keep the shaded region, we must add it back in this expression by adding
A n B n C. This should help you to understand the preceding property.
FIGURE 11
EXAMPLE 33 A survey of 100 first-year students taking social science courses at Lamor
University gave the following information as diagrammed in the given figure:
55 were taking history (H), 45 were taking psychology (P), 25 were taking
sociology (S), 12 were taking history and psychology, 10 were taking history
and sociology, 8 were taking psychology and sociology, and 5 were taking all
three social sciences. Each student is enrolled in at least one of the social
science courses.
r
55
n(H) = 55 P(H) =
1(
n(P) = 45 P(P) =
j^
(S) = 25 jP(5) = "^
n(H n P) = 12 P(H HP) = -j^
n(H n S) = io P(//n S) =
^j
n(pns) = 8 P(Pns) = -nr Fk
100
5
n(H n pns) = 5 P(Hnpns) Too
P(H U P U S) = 1 because each student is enrolled in at least one social-

science course. Also,
P(H U P U S) = P(//) + P(P) + P(S) - P(H DP)- P(H n S)
- P(P n S) + P(H n p n S)
_ 55 ^5_ 25 12 10 8 5
~ + +
100 100 100 100 100 100 100
Check to see if this equals 1>
In the preceding example, find the probability that a student is enrolled in

EXAMPLE 34
history or psychology.
SOLUTION
P(H) =
^ Q
P(P) =
Wn/0 = |
P(H U P) = P(H) + P(P) - P(H n P)
55 45 12 88 22
+ =
~ 100
" 100 100 100 25
Let us now make an observation concerning the probability that an

event does not occur. The probability of getting a 6 on the toss of a die is |.
What is the probability of not getting a 6? There are five equally likely ways
of not getting a 6 namely, getting a 1, 2, 3, 4, or 5. Thus, the probability of

not getting a 6 is |. Note that
5 1
-=l-
6
-
6
In general, let's divide all the events in the sample space into two
mutually exclusive sets A and A' as shown in Figure 13. The set A' is called
the complement of A in relation to the sample space 5.
FIGURE 13
Note that
An A' = and A U A'
Thus,
P(A U A') = P(A) + P(A')

1 = P(A) + P(A')
P(A') = 1 - P(A) or P(A) = 1 - P(A')
Probability of a If A is any event in the sample space S, and if set A' denotes the complement of A, then
Complement
P[A') = '\-P{A) or P[A) = 1-P{A')
EXAMPLE 35 What is the probability of not getting an ace when drawing a card from a
standard deck of cards?
SOLUTION P(no ace) = 1 - P(ace)
= 1-^ = 12
52 13
Sometimes probability statements are given in term of odds, which is

actually a comparison of the probability that event E will occur and that
event E will not occur, E'.
Definition: Odds The odds in favor of event equal
P(E) PIE)
P(E') 1-P()
and the odds against event equal
P(E') 1-P(C)
P() P()
EXAMPLE 36 Find the odds in favor of rolling a 6 with a single die.
SOLUTION
P(6) 1-lW-g
Odds =
P(6) 1/6 _ 1
1 - P(6) 5/6 5
Thus, the odds for rolling a 6 are 1 to 5>
At times we are given the odds for an event and from the odds we
obtain the probability that the event will occur.
Probability from Odds If the odds favoring an event are m to n, then
P() = -r-
m
and P(E') =
y

v
'
m +n '
rr\ + n
EXAMPLE 37 The odds that it will rain today are 1 to 3. What is the probability that it will
rain?
SOLUTION For the given odds, m can be taken as 1 and n as 3.
1 1
P(R) =
1 + 3
Practice Problems 1. If = {1, 2, 3} and = {6, 7}, find

A B B. AU
2. In drawing a card from 8 cards numbered 1 through 8, what is the proba-
bility of getting an even number (A) or a number less than 5 (5)?
3. Using Example 33, find the probability that a student is enrolled in sociol-
ogy but not in history.

4. What are the odds against 2 heads on 2 tosses of a coin?
Exercise Set 5.4 1. If A and B are events with 7. If A and B are events in a sample
Pi A) = .6, P(B) = .3, and space such that P(A) = .6,
P(A n B) = .2, find P(A U B). P(B) = .2, and P(A n B) = .1,
compute each of the following:
2. In Brooks College, 30% of the
(a) P(A')
freshmen failed mathematics,
(b) P(B')
20% failed English, and 15%
(c) P(A U B)
failed both mathematics and En-
(d) P(A' U B')
glish. What is the probability
that a freshman failed mathe- 8. If A and B are events with P(A
matics or English? U B) = |, P(A D B) = i, and
P(A') = 5, compute the follow-
3. An experiment consists of toss-
ing:
ing a coin 7 times. Describe in
(a) P(A)
words the complement of each
(b) P{B)
of the following.
(c) P(B')
(a) Getting at least 2 heads
(d) P(A' U B)
(b) Getting 3, 4, or 5 tails
(c) Getting 1 tail 9. Exam. Let P(A D B) = .2,
(d) Getting no heads P[ A =) .6, and P(B) = .5. Then,

P(A' U B') =
4. A number x is selected at ran-
(a) 1
dom from the set of numbers {1,

(b) .3
2, 3, . . . , 8}. What is the proba-
(c) .7
bility that
(d) .8
(a) x is less than 5?
(e) .9
(b) x is even?
(c) x is less than 5 and even? 10. Exam. A card hand selected
(d) x is less than 5 or is a 7? from a standard deck consists of
2 kings, queen, jack, and one
1 1
5. A single card is drawn from a 52-
10. Three additional cards are
card deck. What is the probabil-
selected at random and without
ity that it is
replacement from the remaining
(a) Either a heart or a club?
cards in the deck. What is the
(b) Either a heart or a king?
probability that the enlarged
(c) Not a jack?
hand contains at least 3 kings?
(d) Either red or black?
3
(a)
6. From a bag containing 6 red 1081
balls, 4 black balls, and 3 green 132
(b)
balls, 1 ball is drawn. What is the 1081
probability that it is 135
(c)
(a) Red or black? 1081
(b) Red or black or green? 264
(c) Not black? (d)
1081
(d) Not red or not black? 267
(e)
1081
Applications
(Business and 11. Marketing Survey. A recent 10% expect higher interest
Economics) survey found that 60% of the rates and a recession
people in a given community 3% expect higher interest
drink Lola Cola and 40% drink rates, higher inflation,
other soft drinks; 15% of the and a recession
people interviewed indicated
What is the probability that a
that they drink both Lola Cola
bank president selected at ran-
and other soft drinks. What per-
dom
centage of the people drink ei-
(a) Would forecast no recession
ther Lola Cola or other soft
or no increase in interest
drinks?
rates?
12. Forecasting. In a survey of the (b) Would forecast no increase
presidents of leading banks by in inflation and no increase
an economics consulting group, in interest rates?
the following information was (c) Would forecastno reces-

obtained relative to their fore- sion, or no increase in inter-
cast for next year: est rates, or no increase in
inflation?
65% expect higher inflation
15% expect a recession 13. Advertising. In St. Petersburg,
5% expect both higher infla- 600,000 people read newspaper
tion and a recession A, 450,000 read newspaper B,
75% expect higher interest and 160,000 read both newspa-
rates pers. How many read either
50% expect higher interest newspaper A or newspaper B?
rates and higher inflation
and Life Sciences) 14. Prediction Relative to Children
and Divorce. In a survey, fami-
lies were classified as C, chil-
dren, and C", no children. At the
same time, families were classi-
fied according to D, husband
and wife divorced, and D', not
divorced. Out of 200 families
surveyed, the following results
were obtained.
15. Survey of Family Characteris-

tics. In a survey of 100 families
of a school district in 1980, each
family was asked the following
questions:
1 Do you have children in pub-

lic school?
2. Do you object to the modern
approach of teaching mathe-
matics?
3. Do you object to placing stu-
dents in classes according to
IQ tests?
The yes answers to these ques-

tionswere tabulated as follows,
where A, B, and C represent
questions (1), (2), and (3), re-
spectively:
n(A) = 70 n(A HB)= 15
n{B) = 30 n(A n C) = 6
n(C) = 10 n(B D C) = 8
n(A n B n C) = 5
(a) If a family is selected at ran-

dom, what is the probability
that the answer from this
family was yes to either
question 1 or question 2?
(b) If a family is selected at ran-
dom, what is the probability
to Practice t> 1.
Conditional Probability and Independent Events 297
5.5 Conditional Probability and Independent Events
Overview When dealing with an uncertain situation, we might expect that as more information is obtained,
the probabilities would change. Alternatively, we might say that as more information is available,
the sample space is modified. Suppose, for instance, that the top executives for the PG&Y
Corporation are evaluating their chances of obtaining a large fabrication contract. They feel that
PG&Y, the Gomez Corporation, and Leary Enterprises are equally likely to win the bidding.
Hence, in their minds, the probability is \ that they will win the contract. Then comes information
that Gomez has withdrawn from the bidding. Excitement reigns at PG&Y because in this modi-
fied sample space, the probability of success for PG&Y is reevaluated at I Could this be true?
You will be able to answer this question when you study
Conditional probability
P(AnB)
[
P[A\B)=
P(B)
Independent events
P{Ar\B) = P(A)-P(B)
Conditional probability is the mathematical term used to describe prob-

ability with additional information. The symbol P(A \B) denotes the problem
of finding the probability that event A will occur, given the information or
condition that event B has occurred; P(A\B) is read "the probability of A,
given B." We consider in this section two procedures for computing condi-
tional probability. In the first procedure, the information is used to obtain a
new sample space reflecting the fact that event B has occurred. This proce-
dure is demonstrated by the next four examples.
EXAMPLE 38 In a sample of 120 students, suppose 80 are enrolled in English, 60 in mathe-

matics, and 20 in both English and mathematics (see Figure 14). What is the
probability that a student selected at random is enrolled in English, P{E)?
What is the probability that a student selected at random is enrolled in
English, given that the same student is enrolled in mathematics, P(E\M)1
FIGURE 14
The given condition of being enrolled in mathematics reduces the number of

possibilities to 60, of which 20 are enrolled in English; thus,
20
P(E\M)
60
EXAMPLE 39 A poll is taken todetermine whether 700 hourly employees of a company

favor a strike. The 700 employees are divided into 3 groups, X, Y, and Z,
according to compensation levels (see Table 5). Suppose an hourly employee

is selected at random. The probability that he or she is in favor of a strike is
ffro. Now suppose an hourly employee is selected at random from group X.
What is the probability that he or she is in favor of a strike?
TABLE 5
In Favor of Not in Favor No

a Strike of a Strike Opinion Total
Group X
^
Then show that (e) P(A\C) = ^JlP
P(C)
and (f) P(D\B) =
P(B)
SOLUTION
P(B n A') is the

- + P(B n A')
probability under A'
in the first row.
P(B n A') = -^
Circled
Likewise, using the first column,
P(A n B') is the

^ + P(AH B') =
probability under A
3
in the second row.
p(A n B')
Circled
P(B') 1 - P(B) Circled at the end

of the second row
1 3
= 1
-
Now, using the information we have obtained in the second row,
\ + W n A') =
\
P(B' n A')
probability under A'
in the second row.
is the
P(B' n A') =
12 Ci rcled]
To check, add the probabilities in the A' column.
, . 1 7 = 2 Correct
Check: T2
+
T2 3
Thus,
F(A|5) = P(A n B) = Ve
=
"
2
/>() ~ 1/4 3
P(fl|A) = P(A n g) _ vfe _

~
l
P(A) ~ !/3 2
= P(A' n B') n
_ /}}_1
P(A'\B')
P(B') ~ 3/4 ~ 9
= />(A' n 5') V12 7

P(B'\A')
/>(A') % 8
In the formula for P(A\B) and P(B\A), if we multiply by the denomina-

tors, we get
P(A n B) = P{B) P{A\B) and P(A D B) = P(A) P(fl|A)

Multiplication Rule The probability that both of two events will occur is equal to the probability that the first
event will occur multiplied by the conditional probability that the second event will occur
when it is known that the first event has occurred:
P(Ar\B) = P(A)P(B\A)
P[AnB) = P{B)-P(A\B)
We quickly observe that this rule gives a procedure for computing the
probability of A and B, something that has been missing from our repertoire
of skills. However, it should be noted that this relationship is helpful only if
one of the relevant conditional probabilities is known or can be computed.
EXAMPLE 42 A basket contains 2 red balls and 2 white balls. A ball is drawn and its color is
noted. Then a second ball is drawn. What is the probability that both balls
are red?
SOLUTION Let /?i be the event of drawing a red ball on the first draw. Then
P(R0 =
^
To find P(R 2 \R\), where R 2 represents a red ball on the second draw, we
consider only the outcomes after R\ has occurred. Since the red ball has not
been replaced, there are 3 balls in the basket, and 1 of these is red. Thus,
P(R 2 \Ri) =
|
Substituting these values in the appropriate multiplication rule gives
PiRi n R2 = ) P(R{) P(R 2 \R0 = \-\ = \
The product formula may be extended to the occurrence of three

events in the following manner:
P(A n B n C) = P(A) P(B\A) P(C\A n B)
Likewise, we can extend this argument to any number k of events, as

indicated in the following formula:
P{A\ nA n...nA ; {)
= P(A t ) P{A 2 \A0 >
/ (A 3 |A, n A-d . . .
P(A k \A ]
D A2 D . . . fl Ak- t)
If knowing B has occurred does not

any additional information yield
about the occurrence or nonoccurrence of A, we say event A is independent
of event B. In this instance, P(A n B) = P(B) P(A\B) = P(B) P(A). We can
use the relationship P(A D B) = P{A) P(B) in situations in which two

events are performed and what happens on the it is clear that has no
first trial
influence on what occurs on the second trial.
Independent Events If two events, A and 6, are independent, then
P{A n fl) = P(A) P{B)
EXAMPLE 43 A card is drawn from a deck of cards. Then the card is replaced, the deck is
reshuffled, and a second card is drawn. What is the probability of an ace on

the first draw and a king on the second?
SOLUTION P(A, n K = 2) P(Ai)

P(K2 \A{).
However, knowing that an ace is drawn on the first draw yields no informa-
tion about what occurs on the second draw, since the first card is replaced
and the deck is reshuffled. Hence,
P(K2 \A X)
= P(K2 )
P(A, n K =P(A )- P(K

2 ) i 2)
_
~
'
52 52
= .0059
EXAMPLE 44 A basket contains 2 red balls and 2 white balls. A ball is drawn, inspected,
and returned to the box. Then a second ball is randomly drawn. What is the
probability of drawing 2 red balls?
SOLUTION Let Ri represent getting a red ball on the first draw and R 2 repre-
sent getting a red ball on the second draw. Since the ball was returned after
the first draw, R {
and R 2 are independent events.
/>(/?, n R2 = ) />(/?,) P(R 2 ) = - - = -l

A A
The theorem on the probability of the occurrence of two independent

events can be extended as follows:
m, n a2 n . . . n A k ) = m,) p(a 2 ) . . . P(A k )

where A\ , A 2 , A 3 , . . . , Ak are independent.
Practice Problems > 1. A class has 10 boys and 2 girls. If 3 students are selected at random, what
is the probability that be boys? all will
2. If you toss a coin and draw a card from a deck of cards, what is the
probability of getting a head and drawing an ace?

Exercise Set 5.5 1. A single card is drawn at random

from a standard deck. Let B =
{the card is black}, H= {the card
is a heart}, and C= {the card is a
club}.
(a) Describe in words a sample
space for the experiment.
(b) How is the sample space
changed if we have the addi-
tional information that a
black card is drawn?
(c) Compute P(H\B).
(d) Compute P(C|B).
(e) Compute P(B\C).
2. Roll a single fair die. Let A =
{the die shows less than 4} and
B = {the die shows an odd num-
ber}. Compute
(a) P(A\B)
(b) P(B\A)
3. If P(A) = .6, P(B\A) = .7, and

P(B) = .6, compute
(a) P(A n B)
(b) P(A\B)
(c) P(B')
(d) P(A U B)
4. Given the following table, com-

pute the probabilities requested.
(a) If a student failed English, (d) What is the probability that

what is the probability that he eats a chocolate and a
she failed mathematics? butterscotch?
(b) If a student failed mathemat-
ics, what is the probability 13. Suppose that the small boy of
that he failed English? Exercise 12 is caught by his
(c) What is the probability that a mother immediately after he
student failed mathematics snatches his Shefirst piece.
or English? makes him return the candy to
(d) If a student did not fail math- the jar. He waits an appropriate
ematics, what is the proba- length of time and then again
bility that he failed English? snatches a piece.
(e) If a student did not fail En- (a) What is the probability that
glish, what is the probability the frustrated thief snatches
that she did not fail mathe- a peppermint, then a choco-
matics? late?
11. A box contains the following he gets chocolate on both
balls: 5 colored red and white, 3 tries?
black and white, 4 green and
white, 6 red and black, 4 red and 14. Exam. An urn contains 10 red,
green, and 5 black and green. 20 white, and 30 blue balls. What
(a) Given that you have drawn a is the probability that, of 6 balls
ball that is partly green, drawn at random with replace-
what is the probability that it ment, 1 ball will be red, 2 white,
is partly white? and 3 blue?
(b) Given that the ball you have 5_
(a) (c)
drawn is partly white, what 36
is the probability that it is 100
(b) (d)
partly red? 649 -n*
12. A candy jar contains 6 pieces of 15. Exam. What is the probability
peppermint, 4 pieces of choco- that a 3-card hand drawn at ran-
late, and 12 pieces of butter- dom and without replacement
scotch candy. A small boy from an ordinary deck consists
reaches into the jar, snatches a entirely of black cards?
piece, and eats it rapidly. He re- 1 . 3
(a) (d)
peats this act quickly. 17
he eats a peppermint and (e)
then a chocolate?
he eats 2 chocolates?
he eats a chocolate and then
a butterscotch?
Applications
(Business and 16. Investments. Of 100 business- that the probability that A is de-
Economics) people polled, 50 have invest- fective is .01, and the probability
ments in common stocks, 35 that B is defective is .05. (Hint:
have investments in bonds, and Since A
and B are fabricated in
25 have investments in both different plants, whether A is
stocks and bonds. What is the good or defective is independent
probability that a person chosen of the quality of B.)
at random from the businesspeo- (a) What is the probability that
ple polled both components are defec-
(a) Invests in common stocks tive?
and not in bonds? (b) What is the probability that
(b) Invests in bonds and not in both components are good?
common stocks?
(c) Does not invest in stocks or
18. Quality Control. You know that
does not invest bonds?in

4% of all light bulbs produced by
(d) Invests in stocks or bonds?
a given company weigh less than
(e) Invests in stocks, if you

specifications, and 2% of all
bulbs are both defective and
know she invests in bonds?
weigh less than specifications.
(f) Invests in bonds, if you
What is the probability that a
know he invests in stocks?
light bulb selected at random is
(g) Invests in stocks, if you
know he does defective, if you know it weighs
not invest in
less than specifications?
bonds?
(h) Invests bonds, if you
in 19. Quality Control. A manufac-
know she does not invest in turer receives a shipment of 20
stocks? articles. Unknown to him, 6 are
17. Quality Control. A
machine is
defective. He selects 2 articles
assembled using components A
at random and inspects them.
and B. The two components are
What is the probability that the
first is defective and the second
built in separate fabricating
is satisfactory?
plants. Experience indicates
and Life Sciences) 20. Genetics. In a study of genet-
ics, a class used a sample of 100
people to obtain the following
information:
21. Genetics. According to the ge- spring being type AA, type aa,
netic theories ofMendel, a par- and type Aa.
ent with genes of type AA can (a) AA mates with AA
transmit only an A gene to off- (b) aa mates with aa

spring. A parent with type aa
can transmit only an a gene.
r
^ ^^ An animal
...
Bg
...
genes is crossed with one with
,
^
However, a parent of type Aa
.
.
D ,
c ., ...
Bb genes. Suppose there is a lit-
.
.
can transmit either an A or an a

..... .
.
r . ,.
ter of 4. What
...
is the
,
UVt
probability
gene, each with probability 5. ,
For each of the following mates,

find the probability of the off-
JM
(b)
wjn bg
A] , wi be
^
flfi?
?
Answers to Practice E> l . /'(all boys)

10 11
Problems
1
2. P(// n A) = P{H) P(A) = l '
2 13 26
Decision-Tree Analysis
A decision-tree diagram allows us to meaningfully arrange the elements of a

complex decision problem without the restrictions of a tabular format. A
further advantage of the decision tree is that it serves as an excellent man-
agement-communication tool, because the tree clearly delineates every po-
tential course of action and all possible outcomes.
The president of Ponderosa Record Company, a small, independent 1
recording studio, has just signed a contract with a four-person rock group
called the Fluid Mechanics. A tape has been cut, and Ponderosa must decide
whether or not to market the recording. If the record is to be test marketed,
then a 5000-record run will be made and promoted regionally; this may result
in a later decision to distribute an additional 45,000 records nationally, for
which a second pressing will have to be made. If immediate national market-
ing is chosen, a pressing of 50,000 records will be made. Regardless of the
test-market results, the president may decide either to enter or not to enter
the national market.
A Ponderosa record is either a complete success or a failure in its
market. A recording is successful if all records pressed are sold; the sales of
a failure are practically nil. Success in a regional market does not guarantee
success nationally, but it is a fairly reliable predictor.
1
Lapin, Lawrence, Quantitative Methods for Business Decisions, 4th ed., 1988, Harcourt
Brace Jovanovich, Inc., Orlando, Florida, pp. 535-537.
The structure of the Ponderosa decision problem is presented in the
decision-tree diagram in the following figure. Decisions are to be made at
two different times, or stages.The immediate choice is to select one of two

acts: "Test market" or "Don't test market." These acts are shown as
branches on the initial fork at node a. If test marketing is chosen, then the
result to be achieved in the test marketplace is uncertain. This is reflected by
an event fork at node b, where the branches represent favorable and unfa-
vorable outcomes. Regardless of which event occurs, a choice must be made
between two new acts: "Market nationally" or "Abort." Each fork corre-
sponds to the two different conditions under which this decision may be
made: at node c, when the test marketing is favorable, and at node d, when it
is unfavorable. If national marketing is chosen at either node c or node d, the
success or failure of the recording still remains unknown, and the possible
events are reflected on the decision tree as branches on the terminal event
forks at nodes /and g.
If the initial choice at decision point a is "don't test market," then a
further choice must be made at the act fork represented by decision point e:
"market nationally" or "abort." As before, node h reflects the two uncer-

tain events that will arise from the branch
a diagrammatical convenience
that allows event and act forks of similar form to appear at the same stage of
the problem and permits all paths to terminate at a common stage. Thus, all
"abort" acts are followed by a dummy branch.
Every path from the base of the decision tree leads to a terminal posi-
tion corresponding to a decision outcome. Each possible combination of acts
and events, or each path, has a distinct outcome. For instance, 0\ represents
the following sequence of events and acts: "Test market," "Favorable,"
"Market nationally," "Success." The real value of the decision tree cannot
be appreciated until we study expected value in the next chapter.
Review the following terms to ensure that you understand their application
to probability:
Important Experiment (257)

Terms Outcome (257)
Sample space (258)
Relative-frequency interpretation of probability (267)
Probability rule on a sample space (259)
Event (260)
Compound event (260)
Probability of an event (261)
A U B (A or B) (286)
A fl B (A and B) (286)
Complement (286)
Mutually exclusive events (288)
Conditional probability (297)
Uniform sample space (260)
Independent events (302)
Simple event (260)
Review the following concepts used in counting:

Important Tree diagrams (269)

Concepts Fundamental principle of counting (270)
Permutations (277)
Combinations (279)
Permutations of n things taken rata time (278)
Combinations of n things taken r at a time (280)
Success in solving probability problems depends on the ability to use

the correct formula. Do you know the conditions that allowyou to use each
of the following formulas?
Important number of times A occurs P(A U B) = P(A) + P(B)

P(A) =
Formulas N P(A UB) = P(A) + P(B) - P(A n B)
< P(A) < 1
P(A') = 1 - P(A)
1
P(A)
n(S) P(A\B) -
p(B)
n(A)
P(A) = P(A n B) = P(A) P(B\A)
n(S)
- - 3-2-1
P(A HB) = P(B) P(A\B)
n\ = n(n l)(/i 2)

P(A n B) = P(A) P(B)

P(n, r)
(n r)
P(n, n) = n\
C(n, r)
r\{n - r)\
Permutations
A?|!/)2-
Review Exercises 1. A number x is selected at ran- (c) x is even?

dom from the set of numbers (d) x > 2?
... (e) x is less than 5 and even?
{1, 2, 3, , 10}
(f) x < 7?
What is the probability that
(g) x is less than 5 or even?
(a) x is less than 5?
(h) x 2 > 3?
(b) x is divisible by 3?
2. A bag contains 6 red balls, 4 rubber galoshes, and 24 contain

black balls, and 3 green balls. boxes of Kleenex.
Tabulate a sample space for (a) What is the probability of
each of the following experi- getting a tape player or an
ments. electric razor?
(a) A single ball is drawn (a (b) What is the probability of
sample space with 3 out- getting something other than
comes). the Kleenex?
(b) A single ball is drawn (a uni-
Suppose P(A) =
6. .35, P(B) = .51,
form sample space).
and P(A n B) = .17. Compute
(c) A ball is drawn and pock-
(a) P(A')
eted. A second ball is (b) P(A U B)
drawn. (A tree diagram
might be helpful.) 7. A die is tossed 3 times. If the
(d) A drawn, its color is
ball is throws are assumed to be inde-
recorded, and it is replaced. pendent, what is the probability
A second ball is drawn. (Try that, on the first throw, the die
a tree diagram.) will show an even number; on
the second throw, a number di-
3. A ball is drawn from the bag in
visible by 3; and on the third
Exercise 2.What is the probabil-
throw, a 1?
ity that it is
(a) Red or black? 8. If two cards are drawn without

(b) Blue? replacement from a deck of
(c) Red or black or green? cards, what is the probability
(d) Not black? that
(e) Not red or not black? (a) Both cards are red?
(f Not red or not green? (b) The two cards are the king
of spades and then the queen
4. A ball is drawn from the bag in
of hearts?
Exercise 2, its color is recorded,
(c) Both cards will be aces?
and the ball is replaced. A sec-
ond ball is drawn. What is the 9. Suppose that in the decision-tree
probability of diagram shown in the Extended
(a) Two red balls? Application on page 307 the
(b) A red ball followed by a probability of favorably test
green ball? marketing is .50, and unfavor-
(c) Two black balls? ably .50. With no test marketing,
the chance of national success is
5. At the Fresh-Air Farm Charity
.60. With test marketing and a
Bazaar, one booth sells 60 iden-
favorable test, the chance of na-
ticallywrapped surprise pack-
tional success is .80. With an un-
ages, where 8 of the packages
favorable test, the probability of
contain tape players, 12 of the
national success is .20. What is
packages contain electric ra-
the probability of favorable test
zors, 16 of the packages contain
marketing and national success?
Chapter Test 311
10. In Exercise 9, what is the proba- (a) All hearts?

bility of unfavorable test market- (b) First 6 red cards and then 7
ing and national success? black cards?
(c) All face cards (jack, king,
11. In Exercise 9, what is the proba-
queen)?
bility of favorable test marketing
(d) First 4 hearts, then 6 dia-
and national failure?
monds, and then 3 other
12. In Exercise 9, what is the proba- cards?
bility of failure in national mar-
15. A shipment contains 96 good
keting if there is no test market?
items and 4 defective items.
13. The and r are writ-
letters/, o, u, Three items are drawn, one at a
ten on four cards and placed in a time without replacement, from
hat. They are drawn one at a the shipment.
time and arranged according to (a) What is the probability of 3
the order drawn. defective items?
(a) Count the number of possi- (b) What is the probability of no
ble arrangements (without defective items?
tabulating them). (c) What is the probability of 2
(b) Tabulate all possible ar- good items and 1 defective
rangements of these letters. item?
(c) What is the probability that (d) What is the probability in
the arrangement drawn will part (a) if the items are re-
have the two vowels adja- placed after each drawing?
cent to one another? (e) What is the probability in
part (b) if the items are re-
14. A 13-card bridge hand is drawn
placed after each drawing?
from a standard deck. What is
the probability that it consists of
Chapter Test
1. What is the probability of getting heads when a 2-headed coin is tossed?
2. In how many ways can 6 books be arranged on a shelf?
3. A box contains 3 red balls and 4 white balls. What is the probability of
drawing 2 white balls
(a) If the first ball is replaced before the second one is drawn?
(b) If the first ball is not replaced?
4. From a group of 7 people, how many committees of 4 can be selected?
5. In a certain college, 30% of the students failed mathematics, 20% failed

English, and 15% failed both mathematics and English. If a student
failed English, what is the probability that he failed mathematics?
6. A file contains 20 good sales contracts and 5 canceled contracts. In how

many ways can 4 good contracts and 2 canceled contracts be selected?
7. If you toss 2 dice, what is the probability of getting a sum of 7 or 8?
8. A box contains 6 red and 4 black balls. Three balls are drawn at random.
What is the probability of getting 2 red balls and 1 black ball?
9. In how many ways can a chairman, a treasurer, and a secretary be

selected from a board of 12 persons?
10. Consider a family of 3 children. Find the probability that all 3 children
are the same sex.
11. The probability that John will live at least 20 more years is 5. The
probability that Marie will live at least 20 more years is i. Find the
probability that neither will live at least 20 more years.
12. If P(A) = .70, P(B) = .30, and P(A D B) = .20, find P(B' U A).
13. A proposal to accept or reject a union is submitted to all employees of

the Arnold Corporation. Of the employees, 30% are laborers, 35% are
white-collar workers, and 35% are blue-collar workers. In the response,
20% of the laborers, 85% of the white collar workers, and 90% of the
blue-collar workers vote to reject the union. What is the probability that
a laborer selected at random voted against the union?
Additional Topics of
Probability
The topics of probability introduced in the preceding chapter are used in this
chapter to solve other types of problems and to introduce special applica-
tions of probability. In the section on finite stochastic processes, special
attention is given to the mutually exclusive ways events can occur in order
to find the probability of a compound event. Problems involving the condi-
tional probability of an event, which depends on the occurrence of an earlier
event, were considered in the last section of the preceding chapter. In this
chapter we reverse the problem and try to find the probability of an earlier
event conditioned on the occurrence of a later one. You will no doubt be
surprised at the number of applications of this type of probability (solved by
Bayes' formula). Bernoulli (or binomial) probability is especially useful in
genetics, biomedical research, and quality control. Random variables and
probability distributions are introduced in order to correctly define a much-
used term of probability, expectation, or expected value. At the end of the
chapter, we introduce Markov chains, in which matrix theory and probabil-
ity are combined to discuss physical systems and how they evolve over time.
6.1 Finite Stochastic Processes
Overview From a box containing 6 red balls and 4 black balls, 2 balls are drawn. What is the probability of
getting a red ball and a black ball?
Note that the problem did not ask for the probability of a red ball followed by a black ball.
In this section we introduce a type of probability problem that we classify as a finite stochastic
313
314 Chapter 6 Additional Topics of Probability
process. The term "stochastic" is derived from a Greek word meaning "guess." In this section
we wish to analyze experiments in which probabilities for each outcome need not be fixed but
take on different values based on what events took place before the trial under consideration.
Applications of finite stochastic processes extend all the way from the genetic theories of
Mendel to quality-control techniques in manufacturing. We make much use of tree diagrams in
this section, and this procedure should be helpful in the application of an important formula in
the next section.
In the preceding chapter we observed that the probability that both of

two events occur in a given order is equal to the probability that the first
will
event will occur multiplied by the conditional probability that the second
event will occur when it is known that the first event has occurred, or
P{A C\B) = P(A) P(B\A)
P(A r\B) = P(B) P(A\B)
In the first probability, A occurs and then B\ in the second probability, B

occurs and then A. It is somewhat easier to understand these formulas by
denoting with subscripts the order in which events occur.
P(A n B 2 = P(A
t ) { ) P(B 2 \A X )
P(A 2 n B { ) = P(B X ) P{A 2 \B X )
In general, since A and B can occur in either order,
P(A nfl) = P(A X

) P(B 2 \A l ) + />(,) P(A 2 \B X )
EXAMPLE 1 From a box containing 6 red balls and 4 black balls, 2 balls are drawn. (This
is equivalent to drawing 2 balls one at a time without replacement.) What is
the probability of getting a red ball and a black ball?
solution This event can happen in two mutually exclusive ways: a red ball
followed by a black ball or a black ball followed by a red ball. Thus,
P{R HB) = />(/?,) P{B 2 \R t ) + P(B0 P(R 2 \B,)
~ '
+
10 9 10 9
=
' 90 =
~ 15
Finite Stochastic Processes 315
The tree diagram in Figure shows all possibilities in the drawing of 2

1
balls from the box. Of course the second and third branches give the desired
probability for this problem.
Probability
P(R,) P(R \R 2 t
)
= t%
PiR,) P{B 2 \R t ) = -k
ne,)
p(R 2 \bo = to
PiB,) P{B 2 \B,) = TT5
FIGURE 1
EXAMPLE 2 Box I contains 3 red and 4 black balls; box II contains 4 red and 5 black balls.
A ball is drawn from box I and placed in box II; then a ball is drawn from box
II. What is the probability that the second ball is red?
SOLUTION The first experiment involves drawing a from box I with two
ball
possible outcomes: R t
, a red ball, or B { , a black ball. The second experiment
involves drawing a ball from box II after a ball has been drawn from box I
and placed in box II. Four possibilities exist, as indicated in Figure 2. There
are two mutually exclusive paths for getting a red ball on the second draw: a
red ball on the first draw and a red ball on the second draw, or a black ball on
the first draw and a red ball on the second draw. The probability of a particu-
lar path is the product of the probabilities along the path. Therefore,
P(red ball on the second draw) = P(R t ) P(R 2 \Ri) + P{B^) P(R 2 \B t )
~ '
+ ' ~ 70
7 10 7 10
Probability
P(R,) P(R 2 \R,)

P(R t
) P(B 2 \R,)
P{B,) P(R 2 \B t )
P(B,) P(B 2 \B,)

FIGURE 2
EXAMPLE 3 Two cards are drawn from a standard deck of cards. What is the probability
of getting an ace and a king?
SOLUTION 1 There are two mutually exclusive ways this event can occur. We
can get an ace on the first draw and a king on the second, A n K2 or a king { ,
on the first draw and an ace on the second, K\ fl A 2 We denote this as .
P(A n K) = P[(A, n k u {K n a
2) x 2 )]
= P(A, n k + P(K, n a
2) 2)
= P(Ai) P(K2 \A t ) + POd) p(a2 |a: 1)
_ 4_ _4_ 4_ 4_
~ '
+ '
52 51 52 51
= .012
13
51
SOLUTION 2 By using the theory of combinations, we note that we are choos-

ing 1 ace from 4 and 1 king from 4. At the same time, we are taking 2 cards
from 52.
C(4, 1)
C(4, 1)
P(A n K) =
' '
On your calculator
FIGURE 3
Thus,
/(positive x-ray) = [P(C) /'(positive x-ray |Q]
+ [P(C')
F(positive x-ray|C')]
= .04(.90) + .96(.05)
= .084
EXAMPLE 5 Suppose that, in a shipment of 100 items, 4 are defective. Items from the
shipment are drawn one at a time and tested. The testing will continue until 2
defective items are discovered. What is the probability that 2 defective items
will be discovered on or before the third draw?
SOLUTION In Figure 4, note that the last path stops when 2 defective items
are discovered. Why? The probability that 2 defective items may be found
on or before the third draw is
P{G t ) P(D 2 \G ]
) P(Z),|G, n D 2) + P(D X ) P{D 2 \D )X
+ />(),)
P(G 2 \D,) P(D 3 |>, n G 2)
_4_ 96 3
+ = .0036
100 99 98 100 99 100 99 98
FIGURE 4
Practice Problems > 1. Box 1 contains 6 red balls and 4 blue balls; box II contains 2 red balls and
6 blue balls; box III A ball is drawn
contains 6 red balls and 2 blue balls.
from Box I. If it is red, itbox II and a ball is drawn from box
is placed in
II. If it is blue, it is placed in box III and a ball is drawn from box III.
What is the probability that a blue ball is drawn on the second draw?
2. Three machines. A, B, and C, produce 50%, 25%, and 25%, respectively,
of all the items produced in a given area of a factory. It has been found
that defective items make up about 5% of the items produced from ma-
chine A, 3% of those from machine B, and 1% from machine C. If an item
is selected out of a day's production from the three machines, what is the
probability that it is defective?
Answers to practice problems arc found on page 321
Exercise Set 6.1 Find the probabilities, given the

probabilities on each branch of the
tree diagram.
9. P(D'\A)
10. P(D\B)
n. P(D'\C)
12. P(A n D)
13. P(B n D')
14. P(C n D)
15. P(A n D')
16. P(A n D)
Find the probabilities, given the
probabilities on each branch of the
17. A new low-flying missile has a
probability of .9 of penetrating
tree diagram.
the enemy defenses and a proba-
bility of .7 of hitting the target if
it penetrates the defenses. What

is the probability that the missile
will penetrate the defenses and
hit the target?
18. Two dice are tossed. What is the (c) One card is odd and the
possibility that the first die other is even?
shows a 3 and the second die an
22. A box contains 2 coins, one of
even number? What is the prob-
which has two tails. A coin is
ability that one die shows a 3
selected at random and tossed.
and the other shows an even
What is the probability of getting
number?
a tail?
19. Two dice are rolled one after the
a head occurs
23. In Exercise 22 if
other. What is the probability
on the first toss, the second coin
that the one die shows an even
is tossed. If a tail occurs on the
number and the other an odd
first toss, the same coin is
number?
tossed. What is the probability
20. Box A contains 3 red chips and 4 of getting a tail on the second
black chips. Box B contains 5 throw?
red chips and 2 black chips. A
24. Urn A contains 5 red and 3 white
chip drawn from box A and
is
marbles; urn B contains 3 red
placed in box B, and then a chip
and 2 white marbles; urn C con-
is drawn from box B. What is
tains 2 red and 1 white marble.
the probability that the chip is
Two coins are tossed. If two
black?
heads appear, a marble is drawn
21. Box A contains 6 cards num- from urn A; if one head appears,
bered 1 through 6, and box B a marble is drawn from urn B; if
contains 4 cards numbered 1 no heads appear, a marble is
through 4. A card is drawn from drawn from urn C. One marble
box A. If it is even, a card is is drawn by this process. What
then drawn from box B; if it is is the probability that it is red?
odd, a second card is drawn
25. Two cards are drawn from a
from box A. What is the proba-
standard deck of cards. What is
bility that
the probability of getting an 8
(a) Both cards are odd?
and 9 in any order?
(b) Both cards are even?
Applications
(Business and 26. Unions. An election to accept ees, 20% of the clerical workers,
Economics) or reject a union is held for all and 5% of the laborers will vote
the employees of the Harrison to reject the union. What is the
Corporation. Of the employees, probability that the union will be
30% are salaried, 35% are cleri- rejected?
cal workers, and 35% are labor-
27. Employment Risks. At CBF En-
ers paid on an hourly basis. Be-
terprises, 60% of the workers
fore the election it is expected
are men. From a personnel
that 80% of the salaried employ-
study it is found that 1 1% of the 29. Quality Control. Out of 10 radio

men and 8% of the women are tubes, 3 are defective. The tubes
"employment risks" because of are tested until 2 defective tubes
health problems or excessive are discovered. What is the
use of alcohol and drugs. What probability that the process
percentage of the employees of stops on
this company are employment
(a) The second test?
risks?
(b) The third test?
(c) The fourth test?
28. Quality Three ma-
Control.
chines, A, B, and C, produce 30. Quality Three manu-
Control.
60%, 30%, and 10%, respec- facturers, and III, supply
I, II,
tively, of all the items produced automobile computers for the

in a section of a factory. It has Hudson Motor Company. Out of
been determined that 6% of the the 20,000 supplied by I, 3% are
items produced by machine A defective; of the 40,000 supplied
are defective, 4% from machine by II, 4% are defective; of the
B are defective, and 2% from 10,000 supplied by III, 6% are
machine C are defective. If an defective. If no further tests are
item is selected out of a day's made and if you purchase a
production from the three ma- Hudson automobile, what is the
chines, what is the probability probability of getting a defective
that it is defective? computer?
and Lite Sciences) 31. Medicine. The probability that on a random selection of adults
a healthy person has a tempera- in this city, what is the expected
ture 0.5 degree above normal is percentage of adults indicated as
.10. The probability that a sick having this disease?
person has a temperature 0.5 de-
33. Mortality. The probability that
gree above normal is .80. Of any
John will live at least 20 more
randomly selected sample of
years is i, and the probability
people, 90% are healthy. What
that Maria will live at least 20
is the probability that a ran-
more years is |. Find the proba-
domly selected individual has a
bility that
temperature 0.5 degree above
normal? (a) Both will live at least 20
more years.
32. Medicine. The probability that
(b) At least one will live at least
a blood test will show a disease
20 more years.
if it exists is .96. The probability
(c) Only John will live at least
that the blood test will indicate
20 more years.
the disease if it does not exist is
(d) Neither will live at least 20
.02. It is believed that 5% of the
more years.
adults in a small city have this
disease. If blood tests are made 34. Genes. An animal with BB
Bayes' Formula 321
genes is crossed with one with 36. Voting. In a certain community

Bb genes. Suppose there is a lit- 55% of those registered to vote
ter of 4. What is the probability are registered as Democrats and
that 45% are registered as Republi-
cans. It is believed that 60% of
(a) All will be Bbl
the Democrats favor the new tax
(b) Two will be BB and 2 B/;'
1
to be voted on, and 30% of the

(c) All will be BB?
Republicans favor the tax. If all
35. Genes. Do Exercise 34 for both registered voters vote, what is
animals having Bb genes. the probability that the tax will

pass?
Answers to Practice 1. P(B) = />(/?i) P(5 : |fl|) + P(B } ) P(B^B^)

Problems P(B) = A-* + A-i = A
2. /ârticle is defective) = P(A) P(D\A) + P(B) P(D\B) + P(C) P(D\C)
particle is defective) = .50(.05) + .25(.03) + .25(.01) = .035
6.2 Bayes* Formula
Overview In this section we are given conditional probabilities in one direction, and need to find condi-
tional probabilities in the opposite direction. That is, P(A\B) may be given or computed, when
the objective is to find P(B\A). Typical problems involve looking at the outcome of an experi-
ment and then asking for the probability that the outcome was due to a particular cause. Such
problems are usually solved by using Bayes' formula (introduced in this section). We introduce
the problem with two boxes, 8, and B 2 each
, containing some red balls and some white balls.
Suppose the probabilities of selecting box 6, and box 8 2 are given. A ball is randomly selected.
If it is red, what is the probability it came from box B, ?
To solve the problem proposed in the overview, let's consider the

following example. Colored balls are distributed in two boxes as follows.
The box contains 3 red balls and
first white ball, while the second box 1
contains 2 white balls and red ball. A box is selected in such a manner that
1
the probability of selecting box B\ is | and the probability of selecting box B 2

is h Figure 5 summarizes these facts.
FIGURE 5
From the first branch of Figure 5, we see that
P{B n R) =
t
-- = -
2 3 1
Also,
1
P(B n R) I Third branch
2 '
3 3
Now,
r = (R n A,) u (R n b 2 )
So
P(R) = P(5, n R) + P(B 2 n R) 5, D R and fl : n /?
are mutually
2 3
I _
U exclusive
~ '
+ I ' "
3 4 3 3 18
From Figure 5 it is easily seen that
P(R nfl,) = j-^
We have learned that
P(R fl ,)
P(B t
\R)
1/2 __9_
11/18
~ 11
So when a red ball is drawn, the probability that it came from B\ is tV-
Now let's consider the same problem without a specific number of red
and white balls and without specific probabilities for drawing each box. The
corresponding tree diagram is seen in Figure 6.
PiR\B t
)
P{W\B 2 )
Bayes' Formula 323
The formula for finding the probability that, if a red ball is drawn, it came
from box B\ can be formulated in the following manner. First note that in
Figure 6
R = (B, n R) U (B n 2 R)
P(R) = P(B\ n R) + P(B 2 n R) 5, D R and B D R

2
are mutually
exclusive
P(B n }
/?) = P(|)
P(/?|#i) First branch
P(B 2 n /?) = P(B 2 ) P(R\B 2 ) Third branch
So
/>(fl) = /'(A,) P(/?|fi|) + P(B 2 ) P(R\B 2 ) Substitution
Also, from Figure 6.
P(R n BO = P(B X
) P(R\B { ) First branch
Now, from Chapter 5,
P(R n bo
Pi BAR) =
PIR)
P(R n b,) Substitution for

P(BO P(R\Bi) + P(B 2 ) P(R\B 2 )
denominator
P(B\)P(R\B\) Substitution for

=
/'(fl,)
/>(/?|j5|) + P(B 2 ) P(R\B 2 )
numerator
In a similar manner, since
m n fl 2 )
= P(B : )

P(R\B 2 ) Third branch
P(B n 7?)
P(B 2 \R) =
2
and
P(R)
P(B 2 ) P(R\B 2
P(B 2 \R) =
)
Substitution
PiBO P(R\B f
) + P(B 2 ) P(R\B 2 )
This interpretation can be extended to three boxes containing red and

white balls:
P(B0 P(R\B t
)
P(Bt\R)
PWt) P(R\B t
) + P(B 2) P(R\B 2 + P(Bi) ) P(R\BJ
Likewise,
P(B 2 ) P(R\B 2 )
P(B 2 \R)
P(B X
) P{R\B<) + P(B 2 ) P(R\B 2 + P(B>) ) P(R\B,)
and
P(B,) P(R\B } )
P(Bi\R)
P[B\) P(R\Bi) + P(B Z ) P(R\B 2 + P(Bi) ) P(R\B y )
In general we have the following formula:
Bayes' Formula Let 6/ (/' = 1, 2 n) with probabilities P{B t ) be a finite set of disjoint events whose
union is the sample space. Let A be an event that has occurred when the experiment was
performed, where A is a subset of the union of the S,. Then
P(B/|/I) =
P{B,)-P(A\B,) + P(B 2 )-P(A\B Z )
+ P(B n )-P(A\B n )
EXAMPLE 6 If P(A\B) = \, P(A\C) = |, P(A\D) P(B) = k, P(C) = |, and P(D) = J, find

P(C\A).
SOLUTION
Bayes' Formula 325
then P will divide from the numerator and denominator, giving the following:
Bayes' Formula, Equally If it is known

Likely Events
(g) P(C) = P(A n + P(B n o o experiment consists of selecting

= P(A) P(C\A) a box (equally likely) and then
+ P(B) P(C\B) randomly drawing a ball from
(h) P(D) the box selected.
(i) P(A\0 =
P(A n q (a) Find the probability of get-
P(C)
ting a red ball.
P(A) P(C\A) (b) Find the probability of get-

P(A) P(C\A) + P(B) P(C\B) ting a white ball.
(j) P(A\D)
(c) Find the probability of get-
(k) P(B\C) ting a green ball.
(1) P(B\D)
(d) If a red ball is drawn, find
the probability that it came
Use the following tree diagram from box I.
and Bayes' formula to find each (e) If a white ball is drawn, find
probability. the probability that it came
from box I.
(f) If a green ball is drawn, find

from box I.
5. Use Exercise 4 to find the fol-

lowing:
(a) If a red ball is drawn, find

from box II.
(a) P(A\D)
(b) If a white ball is drawn, find
(b) P(B\D')
(c) P(C\D)
from box II.
3. Given that = i P(A\B 2

P(A\B t ) )
(c) If a green ball is drawn, find
= i and P(A\Bi) = , find the probability that it came
P(B \A), P(B 2 \A), and PiBÂ)
t
from box II.
under each of the following sets

6. An urn contains three coins; two
of conditions:
of them are fair, and the other
(a) P(B0 = P(B 2 = P(B}) = i ) has heads on both sides. A coin
(b) P(B = \,P(B 2 ) = h,P(B i ) =
]
) is random and tossed
selected at
_5_
twice, coming up heads both
(c) P(B i
)
= lP(B 2 ) = i,P(B i = ) times. What is the probability
2
that the coin selected has two
(d) P(B0 = P(B 2 = I ) P(Bj) = f heads?
4. Box contains 6 red and 4 white

I 7. Box A contains 3 red and 5 black
balls, and box II contains 5 red, chips, box B has 2 red and 3
3 white, and 2 green balls. An black chips, and box C has 1 red
Bayes' Formula 327
and 2 black chips. A box is se- the probability that it came from
lected at random and a chip is box B?
drawn. If the chip is red. what is
Applications
(Business and 8. Salaries. Of the employees of ity that the customer came from
Economics) the JNT Corporation. 5% of the area I? Area III?
men and 7% of the women have
salaries in excess of $40,000.
11. Quality Control. A dealer re-
Furthermore, 60% of the em-

ports a major defect in a TV set.
The set could have been manu-

ployees are men. If an employee
factured any one of three
at
who is selected at random earns
plants. A, B, or C. For the week
more than $40,000, what is the
in which the given set was man-
probability that the employee is
a man?
A made 30% of
ufactured, plant
B made 45%, and C,
the sets.
9. L'nions. A proposal to accept 25%. During this week quality
or reject a union is submitted to control located 2% defects at A,
all employees of the RJW Cor- 3% defects at B, and 5% defects
poration. Of the employees, at C. What is the probability that
30% are laborers, 35% are the TV set came from C? From
white-collar workers, and 35% A?
are blue-collar workers. In the
response, 80% of the laborers, 12. Stock Trends. Three proposed
15% of the white-collar workers, economic projections about the
and 5% of the blue-collar work- future have probabilities of .2,
ers vote to reject the union. .5, and .3 of occurring. If the
What is the probability that a first projection occurs, the prob-
person selected at random who ability that the common stock of
has voted against the union is the Investors Corporation will
(a) A laborer? increase by 5 points is .6; if the
(b) A blue-collar worker? second projection occurs, the

probability of the increase is .7;
10. Marketing. A given area is di-
and if the third projection oc-
vided into submarkets according curs, the probability is .8. Dur-
to the percentage of prospective ing the next year the common
customers: area I is 30%; area stock of the Investors Corpora-
II, 45%; and area III, 25%. From tion does increase by 5 points.
a sampling of the customers, it is What is the probability that the
found in area I that 23% favor a first economic projection has oc-
given detergent. In area II, 20% curred?
favor the detergent, and in area
III, 16%. If a customer is se- 13. Credit. A credit-card company
lected at random and favors the classifies its credit cards in three
detergent, what is the probabil- categories: golden cards ( 10% of

customers), preferred cards each year. The company re-

(30% of customers), and regular ceives a nasty letter from a card-
cards (60% of customers). In the holder whose account was can-
past, 1% of those with golden celed because it had become de-
cards, 6% of the preferred cus- linquent. What is the probability
tomers, and 11% of the regular this person holds a "preferred
cardholders became delinquent card"?
and Life Sciences) 14. Medicine. The probability that person is selected and given the
a person has disease D is P{D) = lie-detector test.
.1. The probability that a medi- (a) If the test indicates he is
cal examination will indicate the guilty, what is the probabil-

disease if a person has the dis- ity that he is guilty?
ease is P{I\D) = and the prob-
.8, (b) If the test indicates he is
ability that the examination will guilty, what is the probabil-

indicate the disease if a person ity that he is innocent?
does not have it is P(I\D') = .02.
What is the probability that
16. Diabetes. If a person has diabe-
Patty has the disease if the medi- tes, a particular blood test
cal examination indicates that shows positive 95% of the time.
she does? On the other hand, the test also

shows positive 2% of the time
15. Lie Detection. Suppose that a for those who do not have diabe-
lie-detector test is 90% accurate tes. About 5% of the general
for guilty people and 95% accu- public have If youdiabetes.
rate for innocent people. If T were to test positive, what
stands for the test indicating would be the probability that
guilt, G stands for a person be- you actually have diabetes?
ing guilty,and / stands for a per-
son being innocent, P(T\G) = 17. Family. A family has 2 girls out
.90 and P(T\I) = .05, or P(T\I) = of 3 children. What is the proba-
.95. One of 5 people in an office bility that the first was a girl
is guilty of stealing money. A given that the last was a girl?
Answers to Practice .4(.2)

1. /'(Cl Profit)
Problems .3(.6) + 3(.3) + .4(.2) 35
.02
2. P(F\D) = -
08 + .02
6.5 Bernoulli Trials and the Binomial Distribution
Overview A new drug is being tested that causes side effects in 6% of the patients. What is the probability
of no side effects if the drug is tested on 20 patients?
Bernoulli Trials and the Binomial Distribution 329
This problem illustrates a general category of probability problems that are concerned
with situations in which an experiment is repeated many times. For example, we might desire to
find the probability of 5 heads in the toss of a coin 10 times, or to find 1 defective item in a
sample of 20. Probability problems of this nature are called problems with repeated trials, or
Bernoulli trials [after Jacob Bernoulli (1654-1705), who contributed significantly to the field of
probability). In each problem, some outcome is designated as a success, and any other result on
a single trial is considered a failure. The probability of success must remain constant from trial
to trial. In order to extend the Bernoulli probability concept to binomial probabilities (or the
binomial distribution), we must first define a random variable and then a probability function of
a random variable.
In this section we are interested in an outcome that can be classified as

a success or as a failure. For example, in tossing a coin either a tail occurs or
does not occur. In tossing a die we either get or do not get a 6. What do these
outcomes have in common? In each experiment there are only two out-
comes: one which we call a success, and its complement, which we call a
failure.
EXAMPLE 9 Suppose we toss a coin 4 times and are interested in the probability of
exactly 2tails. Discuss the relationship between tosses in this experi-
ment.
solution one toss (sometimes called one trial) we either

First note that in
get a tail we get a head (complement of a tail, and called
(called a success) or
a failure). Note also that in the second toss the probability of a tail (namely,
2) is the same as in the first toss. Finally, note that each toss is completely
independent of the other tosses>
This discussion leads to the following definition:
Definition: Bernoulli Trials Repeated trials of an experiment are called Bernoulli trials if
1. There are only two possible outcomes (success or failure) on each trial.
2. The probability of success p remains constant from trial to trial. (The probability of
failure is q = 1 -p.)
3. All trials are independent.
Here are some examples of Bernoulli trials:
1. Tossing a die several times in succession. Consider getting a 6 as a suc-

cess. (Note: p = ,'
is the probability of getting a 6 on one toss, and this
probability remains constant from toss to toss. In addition, the results of
one toss are completely independent of another toss.)
2. Drawing a card from a deck several times, replacing it each time, and
noting whether or not you get a heart.
EXAMPLE 10 Consider the experiment of tossing a die 4 times. Getting a 6 will be a

success. What is the probability that the die will fall: failure, success, suc-
cess, success?
SOLUTION We can write these outcomes as
Fnsnsns or fsss
Now
P(S) = P(6) =
^
and
P(F) = 1 - P(S) = 1
- \ =
I
The trials are independent, so
F(FSSS) = P(F n s n s n S) = F(F) P(S)

P(S)
P(S)
5
=
~ 1
I I
' ' '
-=~
6 6 6 6 1296
EXAMPLE 11 Consider the experiment in Example 10, but this time find P(SFSS).
solution />(SFSS) = P(S n f n s n S)
= P(S)
P(F)
P(S) P(S)
= = _5_
~ [6 I I [
'
6
'
6
'
6
~ 1296
We note from the two preceding examples that, for two of the possibili-
ties for obtaining three 6's out of four tosses of a die, the probability remains
constant as tAb- In fact, the probability of three S's and one F in any order is
tAb. Since these possibilities are mutually exclusive, to find the probability
of three 6's in four tosses, we add the probabilities of the individual possibili-
ties. How many Now
you could get a success on the
possibilities are there?
first toss (Si) or on the second We can have
(S 2 ), third (S 3 ), or the fourth (S 4 ).
only three successes. In how many ways can we select a combination of
three successes from a possibility of four? The answer is the number of
combinations of four things taken three at a time, or C(4, 3). Therefore,
5
F(three 6's in 4 tosses) = C(4, 3) t^kt
= ili ~ JL
" 1296 = 324
In general, the preceding reasoning can be used to validate the follow-

ing theorem:
Probability of x Successes The probability of exactly x successes (and n - x failures) in a sequence of n Bernoulli
in n Bernoulli Trials trials is given by
P(x successes) = C(n, x)p*g"~ x
where p and q are the probabilities of success and failure, respectively.
EXAMPLE 12 An who read their

advertising agency believes that 2 out of every 3 people
advertisement purchase the product involved. Five people who have
will
read the advertisement are selected at random. What is the probability that
exactly 3 will purchase the product?
SOLUTION Using C(, x)p q"- x

x
written as C(n, x). The number of successes in a binomial experiment is
designated as a binomial random variable.

First let's define what we mean by a random variable. When we per-
form probability experiments, such as tossing a coin, rolling a die, and
counting defective items, some real number usually exists for each outcome
of an experiment. This introduces an important idea of probability, the ran-
dom variable. For our purposes, we will consider the following abbreviated
definition of a random variable:
Definition: Random A random variable is a function or rule that assigns numerical values to the elements of a
Variable sample space.
That is, a random variable is not really a variable, but a procedure by

which values are assigned. For a random variable, we usually use a capital
letter, say X.
A random variable is said to be finite if it can take on only a finite
number of different values. The following is an example of a finite random

variable.
EXAMPLE 14 Toss a pair of dice and define the random variable X to take on values that
are the sum of the two numbers on top. What are the possible values of this
random variable?
SOLUTION The values of A' we associate with this experiment are 2, 3, 4,

that
5, 6, 7, 8, 9, 10, 1 1 , and 12, sum of the two numbers on top of the two dice
the
(see Table 1). Since X can take on only a limited number of values (11), it is
said to be finite.
TABLE 1
Variable
is sis, and the probability that a 3 occurs is $g. This introduces the concept of
the probability distribution of a random variable.
Definition: Probability Let X be a finite random variable. The probability distribution function of X is a function
Distribution Function of a or rule p(x,) defined for 1 < / < n by
Random Variable
p(x,) = P(X = x,)
where n is the natural number of elements x,.
The preceding distribution is sometimes called a discrete probability

distribution function. It is discrete because it is defined only for integral
values, 0, 1, 2, 3 n. Graphically, this probability distribution function
is defined by a finite number of points, as shown in Figure 7.
/'(-v.,
pUi
FIGURE 7
EXAMPLE 15 Graph the probability distribution function of the random variable in Exam-
ple 14.
solution The probability distribution function of the random variable x is
shown in Figure 8.
From the definition of a probability distribution function and the pre-

ceding examples, you can discover the following properties of a probability
distribution function:
Properties of a Probability A probability distribution function (sometimes called probability density function) of a
Distribution Function discrete random variable satisfies
1. 0<p(x,)<1 (i = 1,2 n)
2. p(x-i) + p(x 2 + )

+ p(x) = 1 where x-, , x2 x n are all the possible values
assigned by the random variable.

p(x) = P (sum of dice = ,v)
-\ 1 1 -v
9 10 11 12
FIGURE 8
Note in the preceding example that ^ p(.v,) < because eachp(jc,-) is a

1
probability. The sum of all the p(x,) is 1 because the probability of getting
either a 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1 , or 12 in tossing two dice is 1
Let's now consider a discrete probability distribution that represents

the number of successes in /; Bernoulli trials with corresponding probabili-
ties CUu x)p x q"~ x This probability distribution
. is called the binomial distri-
bution.
EXAMPLE 16 Consider again rolling a die 4 times where getting a 6 is a success. Find the
probability of getting x successes, and tabulate the values of the variable and
corresponding probabilities of the discrete probability distribution.
SOLUTION
(\Y(S\
P(x successes in 4 tosses) = C(4, x)[t) (rj
The probability distribution is defined by Table 2.
TABLE 2
X
Practice Problems f> 1. A coin is tossed 6 times. What is the probability of obtaining more than
one head?
2. A family decides to have exactly 4 children. What is the probability of
exactly 2 girls and 2 boys?
Exercise Set 6.3 Find C(n, x)p"q n " for the given val-
ues of n, x, and p.
1.
16. Let .V be the number of suc- (a) P(x = 6)

cesses in 10 independent repeti- (b) P(x = 0)
tions of a Bernoulli experiment (c) P(x > 8)
where the probability of success (d) P(x < 6)
on a single trial is .4. Find each (e) P{x > 9)
of the following: (f P(x < 2)
Applications
(Business and 17. Sales. A company has found (a) Exactly 1 is defective?
Economics) that 25% of all customers con- (b) Exactly 3 are defective?
tacted will buy its product. If 10 (c) At least 1 is defective?
customers are contacted, what (d) Fewer than 3 are defective?
is the probability of more than 2
19. Quality Control. In the manu-
sales?
facture of a certain item, under
18. Quality Control. Suppose that normal conditions 4% of the
10% of the items produced by a items have a defect. If 10 items
factory are defective. If 6 items are selected from an assembly
are chosen at random, what is line, what is the probability that
the probability that exactly 9 will be defective?
and Life Sciences) 20. Fertility Drug. The probability fects in 6% of the patients. What
of multiple births for women us- is the probability of no side ef-
ing a certain fertility drug is fects if the drug is tested on 20
20%. Ten women take the fertil- patients?
ity drug. What is the probability
23. Divorce. The probability that a
of more than 2 but fewer than 6
couple in Colorado will get a di-
of the women having multiple
vorce within the next 10 years is
births?
I. Six couples living in Colorado
21. Drug Effectiveness. A drug form a dinner club. What is the
manufacturer claims that a par- probability that at least one of
ticular drug is effective 90% of the couples in the club will get a
the time. A physician prescribes divorce in the next 10 years?
the drug to 10 patients and 6 re-
24. Politics. Two-thirds of the par-
spond to this treatment. What is
ticipants at a state Republican
the probability of 6 or fewer suc-
convention are conservatives. If
cesses if p = .9? What conclu-
7 members of the convention are
sion is the physician apt to
chosen at random to serve on a
draw?
committee, what is the probabil-
22. Side Effects. A new drug is be- ity that the conservatives will be
ing tested that causes side ef- the majority?
Expected Value and Decision Making 337
Answers to Practice 1. /"(more than I head) = I

- Pino heads) - P(\ head)
Problems
->-'-'-
2. ^(exactly 2 boys and 2 girls) = C(4, 2) (-) (-) = -
6.4 Expected Value and Decision Making
Overview Probabilities by themselves do not always supply all the information that is useful when making
a decision in an uncertain situation. Many decisions in business and science are made on the
basis of the outcomes of specific desires. For example, what a businessperson stands to gain or
lose in a transaction is important in making a decision. The tool used to discuss such expected
gain or loss is expected value. It was originally developed for games of chance, and we will
discuss several games of chance in this section. In addition we will use a formula for finding the
expected value, or expectation, for the binomial distribution discussed in the previous section.
An important property associated with probability is that of expecta-

tion, or expected value. If we toss a coin 100 times, we could expect to get a
head
100 - = 50 times
If we spin a spinner with 10 equal sections 1000 times, we would expect the
spinner to stop on any given section
1000
-jjj
= 100 times
The concept is perhaps most easily explored in the analysis of a simple

game of chance. Suppose some poor benighted soul were persuaded to play
the following game. A fair coin is flipped. If a head appears we receive $5
from our opponent; if a tail appears we pay our demented opponent a sum
of $2.
We investigate what
happens if we play the game 100 times. In 100 flips
of the coin we can
expect approximately 50 heads and 50 tails. Hence, we
can expect a payoff of approximately (50) ($5) from the heads and a payoff of
(50) (-$2) from the tails (we lose $2 each time a tail occurs). Our net profit
would thus be (50) ($5) + (50) (-$2) = $150. Since the game was played 100
times, our average profit per game would be loo = $1.50.
It is very important to observe that there is an alternative way to
compute this average gain per game.
/^winning $5) = P(H)

/'(winning - $2) = P(T)
($5)/>(winning $5) + (-$2)P( winning -$2) = =

$5 + -z -$2
= $2.50 - $1 = $1.50
Definition: Expected If an experiment has a probability distribution

Value
average number of heads is 1.5. This is true even though 1.5 heads never
happens in a single toss.
EXAMPLE 18 A nationwide promotion promises a first prize of $25,000, 2 second prizes of

$5000, and 4 third prizes of $1000. A total of 950,000 persons enter the
lottery.
(a) What is the expected value if the lottery costs nothing to enter?
(b) Is it worth the stamp required to mail the lottery form?
SOLUTION
(a) Since
/>($25,000) = 7^77-^ ^
/>($5000) =
950,000 '
950,000
m,000, =
95^000
4
W > =
949 993
9^000
the expected value is
{$25 - m 95o!o00
+ i$5 m 9^000 + ,$,0 0)
95(0)00
= 041
In repeated participation in this lottery, the average value of prizes won

would be $0.04.
(b) Not when the expected value is only approximately $0.04>
Expected value is games of chance. In a fair game,

useful in studying
the expected value of the game Casinos and lotteries understandably do
is 0.
not operate fair games; instead, they operate games in which the player has
an expected value that is negative. That is, the player can expect to lose
money on repeated playing of the game.
Consider a game in which a player pays $2 for the privilege of playing.
Suppose her probability of winning $10 is T n and her probability of losing is
!
ttt. Actually, if she wins, her winnings are $10 - $2 = $8 because she pays $2
to play. If she loses, she will lose $2. The expected value of the game is
fi(r) = $8 (ij) - $2 Q = -$1
Thus, if the player should continue to play the game, her average earnings
would be -$l. That is, she would lose, on the average, $1 per game.
It should be clear that expected value would be helpful in making
decisions involving uncertainty. One method
choosing a course of action in
is to choose the action that gives the largest expected value. We will see
throughout the remainder of this chapter how expected value plays a role in
monetary situations and in various social-science problems.
EXAMPLE 19 A company has the privilege of bidding on two contracts, A and B. It is

estimated that, if the company should win contract A, a profit of $14,000
(above all costs) would be realized; however, it costs $500 to prepare a

proposal in order to submit a bid. Contract B would give a profit of only
$10,000 (above all costs), but the cost of preparing a proposal would be only
$200. It is estimated that the probability that the company will win contract
A is j, and the probability of winning contract B is g. If the company can
submit a proposal for only one contract, which proposal should be sub-
mitted?
SOLUTION The probability of winning contract A is 1, so the probability of not

winning this contract is 1 \=f . If the contract is won, the profit is $14,000;
if the contract is not won, the loss is $500. Thus,
(A) = $14,000 Q+ (-$500)

(^)
= $3125
Similarly, the expected value from B is
E(B) = $10,000 (|) + (-$200)

(|)
= $3200
If the expected value is considered as an appropriate criterion for a decision,
the proposal should be for contract B>
EXAMPLE 20 Students over the years have observed that Professor Ab Stract is never on
time and yet never misses class. Records kept by students in the back row
indicate the probabilities shown in Table 4.
TABLE 4
Minutes Late
SOLUTION
(a) = 1 Pd) + 2 P(2) + 3 PO) + 4 P(4) + 5 P(5) + 6 P(6)

= l(.l) + 2(A) + 3(.3) + 4(.l) + 5(.05) + 6(.05)
= 2.75 minutes
(b) (45)(2.75) = 123.75 minutes
We now find the expected value of a binomial distribution.
EXAMPLE 21 Consider rolling a die 4 times, where getting a 6 is a success. Find the
expected number of successes.
solution In Example 16, the following table defined the probability distribu-
tion function. The last column will be used to find the expected value.
TABLE 5
X
x (defects) P(x)
C(6, 0)(.001)"(.999)
6
.99402
l
C(6, l)(.001) (.999)

5
= .00597
2
C(6, 2)(.001) (.999)
4
= 00001
3
C(6, 3)(.001) (.999)
3
=
4
C(6, 4)(.001) (.999)
2
=
5
C(6, 5)(.001) (.999)' =
6
C(6, 6)(.001) (.999)
The mean of the distribution is
np = 6(.001) = .006
EXAMPLE 22 The computer failure in the first year of a new Oldsmobile is

probability of a
The Southeast region sold 100,000 Oldsmobiles. Find the expected
.0001.
number of computer failures.
SOLUTION
E(x) = np p = .0001 n = 100.000
= 100,000(.0001)
= 10
Thus, 10 computer failures are expected in the Southeast region.
Practice Problem 1. Alfa Car Insurance Company insures 200,000 cars each year. Records
indicate that during the year the company will make the following liability
payments for accidents.
Liability Corresponding Probabilities
500,000 .0001
100,000 .001
50,000 .004
30,000 .01
5,000 .04
1,000 .06
.8849
What amount could the company expect to pay per car insured?
Answer to practice problem is found on pane 346.

Exercise Sel 6.4 1. An alphabet block (six sides

containing letters A, B, C, D, E
and F) is rolled on the floor 360
times. How many times would
you expect the A to be on the
top?
Find the expected value for each of

the following:
2. Outcome
Find the expected value for the ran- 15. Records show that 0.2% of the
dom variable whose probability taxicabs in New York City have
function is graphed in Exercises accidents each day. If a com-
13-14. pany has 100 cabs operating on a

given day, how many accidents
13. p(x)
can be expected?
.3-
16. Sam Selpeep is noted for his
ability to sleep through an entire
class. Following a psychology
lecture during which he per-
formed admirably by sleeping
throughout the class period, he
was given a 10-question, true-
false test. What is his expected
14.
number of correct answers?
.5
.4-- 17. Find the expected value of the

number of dots that appear in
.3--
the toss of 3 dice.
.2
Applications
(Business and 18. Contracts. The Ronco Corpora- (b) Estimated profit $1000, cost
Economics) tion prepares a bid on a job at a of proposal $200, probability

cost of $7000. They estimate of winning 5
that if they get the job, they will (c) Estimated profit $10,000,
make $250,000 in profits. If the cost of proposal $600, proba-
probability of getting the job is bility of winning 3
.4, what is their expected profit

20. Sales. During a sale, an appli-
or loss?
ance dealer offers a chance on a
19. Contracts. Find the expected $1200 motorcycle for each re-
values for the following con- frigerator sold. If he sells his re-
tracts. frigerators at $25 more than
(a) Estimated profit $5000, cost other dealers, and if he sells 120
of proposal $500, probability units during the sale, what is the
of winning g
expected value of a purchase to 23. Quality Control. The output of

the consumer? an automatic machine at Ander-
son Corporation is analyzed and
21. Sales. A company has found
found to be a binomial process
that 25% of all customers con-
with a probability of a defect of
tacted will buy its product. If 20
.04. Consider a sample of 6 units
customers are contacted, how
to be tested. What is the ex-
many sales can be expected?
pected number of defects?
22. Quality Control. During inspec-
24. Training Program. In a training
tion of 1000 welded joints pro-
program, a company has a drop-
duced by a certain machine, 100
out rate of 12%. If 50 trainees
defective joints are found. Con-
start the program, how many
sider the random variable to be
can the company expect to finish
the number of defective joints
it?
that result when 5 joints are
welded. How many defective
joints are expected?
and Life Sciences) 25. Mortality Table. According to a devises a scheme that costs $1
mortality table, the probability per questionnaire, but she thinks
that a 20-year-old woman will she will get an 80% return.
live 1 year is .994; the probabil- Which scheme should she use
ity that she will die is, of course, for minimum expected cost?
.006. She buys a $1000, 1-year
28. Unemployment. The probability
term life-insurance policy for
that asteelworker will remain
$20. What is the expected loss or
employed during the next year is
gain of the insurance company?
.866. Each steelworker who
(Assume no interest for the
loses his job is eligible for $2000
year.)
in unemployment benefits from
26. Mortality Table. The probabil- his state. How much money
ity that a man, age 40, will live should the state have in its
for 1 year is .906. How large a budget for each steelworker?

premium should he be willing to
pay for a 1-year, $2000 term pol-
29. Testing Techniques. A psychol-
ogy professor notes that a sys-
icy? (Assume no interest for the
tematized review increases her
year.)
students' scores on final exami-
27. Polls. A public-opinion pollster nations. Of the number of times
finds that, in a mailing process she has given final examinations
costing $0.50 per questionnaire, after systematized reviews,
she gets a 40% return. Her fol- scores have increased i of the
low-up procedure to ensure a re- time by 20%, \ of the time by
ply to most of the questionnaires 10%, I of the time by 8%, and \
costs $3 per questionnaire. She of the time by 4%. What is the
expected increase in final-exam- convention are conservatives. If

ination scores if the professor 12 people are chosen at random
gives her students a systema- to be the rules committee, how
tized review? many conservatives can you ex-

pect to be on this committee?
30. Politics. Two-thirds of the par-
ticipants at a state Republican
Answer to Practice > 1. The expected liability per car is $910.

Problem
6.5 Markov Chains
Overview In this section, matrix theory and probability theory are combined to focus on a new application.
In the first section of this chapter we introduced stochastic processes, processes or experi-
ments in which outcomes depend on previous outcomes. If the outcome of an experiment is
dependent only on the outcome immediately preceding it, the process is called a Markov
process, or Markov chain. Andrei Andreyevich Markov (1856-1922) was a Russian mathemati-
cian who first introduced the study of this type of stochastic process. Weather patterns, the
behavior of animals in psychological tests, population studies, the study of price and market
trends, as well as many other processes can be modeled using this technique. In this section we
will

Write transition matrices for Markov-chain problems
Find steady-state matrices for Markov processes

Determine when a Markov chain is regular
The United Way of Atlanta has determined that 90% of those who
contribute one year will contribute the next year. (This of course implies that
10% do not contribute the next year.) Also, 20% of those who do not contrib-
ute one year will contribute the next year. This information can be repre-
sented by what we call a transition matrix, where C represents those who
contribute and DNC represents those who do not contribute.
Next State
C DNC
C [.9 .1
Present State
DNC .2
Note in the preceding discussion that today (called "present state") a

person is classified as a contributor (C) or a noncontributor (DNC) and the
elements of the matrix give the probabilities that these will be contributors or
noncontributors in the next state. In this section we are interested in a
process moving from one state to the next or a transition from the present
state to the next state with the following property:
Markov Chains 347
Definition: Markov Chain If the probabilities of the outcomes of any trial except the first depend on the outcome of
the preceding trial only, the series of trials is called a Markov chain.
EXAMPLE 23 The following matrix represents the various probabilities of smoking or non-
smoking fathers having smoking or nonsmoking sons.
Sons (Next State)
Smoking Nonsmoking
Smoking .2"
Fathers (Present State)
Nonsmoking
We obtain the desired probabilities by locating the appropriate row and

column. For example, the probability of a smoking father having a nonsmok-
ing son is only .2, whereas the probability that a nonsmoking father will have
a nonsmoking son is ,6>
We illustrate a transition matrix of a Markov chain by a simple experi-

ment having only two possible outcomes, 5] and S 2 If the present state is 5,
and the experiment is performed, then the next state can be either S\ or S 2 .
Likewise, if the present state is S 2 then the next state can be S\ or S 2 Note
,
that there are four transition probabilities involved:
Pi, , probability of going from 5, to S\ Staying in S\
Pxi, probability of going from S\ to 52
P:i , probability of going from S2 to S\
p 22 , probability of going from S2 to S2 Staying in S2
The tree diagram in Figure 9(a) and the transition diagram in (b) show
the possibilities of this experiment.
(a)
Pit
FIGURE 9
Thus, the matrix
P\\ Pl2
P2I P22
is called a transition matrix for a Markov chain. Such matrices have the
following characteristics.
Definition: Transition A transition matrix of a Markov chain having n states is the n x n matrix
Matrix
Markov Chains 349
performed four times. The following example illustrates an interesting way

to obtain probabilities of repeated experiments.
EXAMPLE 25 Use the data of Example 24 to find the transition probabilities after 2 free
throws have been shot.
SOLUTION The transition probabilities are listed in Figure where S, repre-

1 1 ,
sents the state of having made a free throw and S 2 represents the state of
having missed one.
FIGURE 11
Note that the transition can be made from state 1 to state 1 in two experi-
ments in two ways: through either 5i or S 2 at the end of one experiment. The
same is true for the other transitions, the probability of each being the sum of
two products.
p u (2) = .80(.80) + .20(.60) = .76 Si to Si
p x2 {2) = .80(.20) + .20(.40) = .24 S, to S2
p 2 ,(2) = .60(.80) + .40(.60) = .72 S2 to S,
p 22 (2) = .60(.20) + .40(.40) = .28 S2 to S2
Let's note what happens when we square the transition matrix A:

2
80 .20l .76 .24
A 2
60 .40 .72 .28
A comparison of the probabilities within the square of the transition matrix

and the transition probabilities after 2 free throws indicates that
A2 _ \pu(2) p n {2)
L/>2.(2) p 22 (2)J
The preceding example illustrates the following theory:
Power of a Transition If a transition matrix of a Markov chain is raised to the nth power, the element Pij(n) of
Matrix the nth power of the matrix gives the probability that an experiment in state S, will be in
state Sy after n repetitions of the experiment.
EXAMPLE 26 Given the transition matrix
suppose an experiment is in state 1. Find the probability it is in state 2
(a) One step later

(b) Two steps later
(c) Three steps later
SOLUTION JL _Z_
12 12
A 2
J_ J_L
A'
IK IX
(a) From A, p 12 (l) = I

(b) From A 2 p 12(2) = h
,
(c) From A-\ p 12(3) = H
We return now to the example at the beginning of this section, where

Markov Chains 351
9 .!
[.634 .366] = |.644 .356] Fourth state
8
|. 644 .356] [.651 .349| Fifth state
[.65 .349] = 1.656 ,344| Sixth state
[.656 .344] .659 .341 Seventh state
Actually, the state probability matrices approach or converge to a fixed

probability matrix:
[.666 .334]
When this happens, the system is said to be in equilibrium; additional repeti-

tion of the transition matrix will not change the state matrix very much.
Thus, we will have found a probability matrix X such that XA = X.
Definition: Fixed or A state matrix X such that

Steady-State Matrix
XA = X
is called a fixed or steady-state matrix for the transition matrix A.
EXAMPLE 27 Find the steady-state probability matrix for the matrix

"0 1
SOLUTION Using the preceding theorem, we are looking for a steady-state

probability matrix X such that XA = X. Suppose we denote X by [x y].
Then
[x v]
X + - y Multiply matrices
[5
- y = .v and x + y = y Set elements equal

[jj
352 Chapter 6 Additional Topics ot Probability
Recall that
x + y = 1 x and y are probabilities

with x + y = 1
Thus,
y + v = 1
-v=1 Add coefficients
Divide by f
3
X = 7T x + y = 1
We can easily verify that [| h] is a steady-state matrix for our transition

matrix as follows:
[0
Markov Chains 353
.28
Properties of Regular
Transition Matrices
since each row of M

is identical to the steady-state probability matrix X.
Note that the long-term probability of being in state 1 is |, regardless of

whether the initial state is 1 or 2. Likewise, the long-term probability of
being in state 2 is |.
Now the steady-state matrix for V = [100 200] is
V M = [100 200]
900 1500
Hence, V A, VA VA 2
,
3
. approaches [^g SH- Note also that for
steady-state matrix V,
VA = V
900 1500 f 1 900 1500 1
3 2
5 5.
Practice Problems > 1. A Markov chain has a matrix

Markov Chains 355
I k
19. Repeat Exercise 18 for the tran-
12. 1
sition matrix
1 '.2
7 .1
13. 5 .2
2 .4
14.
.4
15.
1
16. A Markov chain has the transi-

tion matrix
~.7 .3
(a) What is the probability of

starting in state 1 and going
to state 2?
(b) What is the probability of be-
ing in state 2 and staying in
state 2?
17. A Markov chain has the transi-

tion matrix
3 .3 .4
2 .2 .6
1 .5 .4
(a) What is the probability of

starting in state 1 and going
to state 3?
(b) What is the probability of be-
ing in state 3 and going to
state 2?
(c) What is the probability of be-
ing in state 3 and staying in
state 3?
18. A Markov chain has the follow-

ing transition matrix. Draw a
tree diagram to represent it.
.2 .8"
.7 .3
ri
3
29.
30.
31.
32.
Markov Chains 357
(c) What is the probability of be- (d) What is the probability of be-
ing in state 2 and being in ing in state 1 and remaining

state 1 three steps later? in state 1 three steps later?
Applications
(Business and 52. Marketing. Two newspapers in (b) If a man is a blue-collar
Economics) town, the Star and the Times, worker, what is the probabil-
are competing for customers. A ity that his grandson is also a
study shows the following: At blue-collar worker?
the beginning of the study, the (c) If a man is a blue-collar
Star had of the customers; at worker, what is the probabil-

the end of the year, it is found ity that his great-grandson is
that the Star kept60% of its cus- a white-collar worker?

tomers and lost 40% to the 54. Advertising. After an intensive
Times. The Times kept 70% of advertising campaign, it was
its customers and lost 30% to the
found that 90% of the people
Star. What will be the customer continued to use product X and
distribution
10% changed brands. Of those
(a) At the end of next year? not using product X, it was
(b) Two years after the study
found that 40% switched to
was made? product X and 60% continued to
(c) What is the long-range distri-
use what they had been using.
bution prediction?
(a) Write a transition matrix for
53. Occupational Probabilities. The
this Markov chain.
following transition matrix gives
(b) After 2 intensive advertising
occupational-change probabili-
campaigns, what percentage
ties:
Sons of people changed to prod-
White Blue uct X from other brands?
collar collar (c) After 2 campaigns, what per-
White collar |
centage of those using prod-
Blue collar uct X decided not to use
product X?
(a) If the father is a blue-collar
worker, what is the probabil- 55. Demographics. Over a given
ity that a son is a white-col- year, the following shifts in em-
lar worker? ployment are recorded:
Employment Shift Percentage
From business (or industry) to business (or industry) 75

From business (or industry) to unemployment 15
From business (or industry) to self-employment 10
From unemployment to business (or industry) 25

From unemployment to unemployment 60
From unemployment to self-employment 15
Employment Shift Percentage

From self-employment to business (or industry) 10
From self-employment to unemployment 10
From self-employment to self-employment 80
How will people be distributed

in the various employment cate-
gories in the long run?
and Life Sciences) 56. Genetics. A basic assumption
in a simple genetics problem is
1 1
that the offspring inherits one 2 2
gene from each parent and that

these genes are selected at ran-
dom. Suppose an inheritance
trait is governed by a pair of
genes, each of which is of type G

or g. The possible combinations
are gg, gG (same genetically as
Gg), and GG. Those possessing
the gg combination are called re-
cessive (denoted by R); hybrid
(denoted by H) is used to indi-
cate those possessing gG; and
dominant (denoted by D) indi-
cates those possessing GG.
Note that in the mating of two
dominant parents, the offspring
must be dominant. In the mating
of two recessive parents, the off-
spring must be recessive. In the
mating of a dominant parent
with a recessive parent, the off-
spring must be hybrid. Suppose
a person of unknown genetic
character is crossed with a hy-
brid, the offspring is again
crossed with a hybrid, and the
process is continued, hence
forming a Markov chain; the
transition matrix is
Absorbing Markov Chains 359
state. Under certain conditions, a stochastic process will give rise to a transition matrix for a
Markov chain with the property that once a given state is reached it is impossible to move out of
that state. Such Markov chains are called absorbing Markov chains and these chains are useful
tools in many fields of study; they are especially useful in the designing of models by research
scientists. In this section we will
Learn to identify absorbing states and absorbing Markov chains

Find the limiting matrix of an absorbing Markov chain and interpret the elements
Determine the expected number of times an absorbing chain will visit a transient state
Consider the transition matrix

r i
4
_
i
EXAMPLE 32 Show that
1
EXAMPLE 34 Write in canonical form the transition matrix of the following absorbing
Markov chain.
EXAMPLE 36 Partition the canonical form of

2. The elements of FR are the probabilities of nonabsorbing states moving to absorbing
states.
3. The sum of the entries in the /th row of the fundamental matrix gives the expected
number of transitions of a system that begins in the /th state before reaching an
absorbing state.
4. The entry in the /',

/
position of the fundamental matrix gives the expected number of
times that a system that begins in the /th nonabsorbing state will be in the /th nonab-
sorbing state before reaching an absorbing state.
To find the limiting form of the transition matrix
2
and second absorbing states, ff and f| are the probabilities that the
first last
nonabsorbing state will go into the first and second absorbing states.
EXAMPLE 38 Consider the following maze:
-II-
FIGURE 13
A mouse is just as likely to leave a state as to stay in the state. The mouse
can enter state 4 but cannot get out. Set up a transition matrix and find the
expected number of transitions, or states the mouse will occupy, with the
mouse starting in each state before reaching absorbing state 4. If the mouse
is released in state 2, in state 1, or in state 3, find the expected number of
times it will be in each of the nonabsorbing states.
SOLUTION The probability of staying in state 1 is f the probability of leaving

;
is 2. The probability of going to state 2 or state 3 is then \. Therefore, we

have the first line of the transition matrix. Can you verify the remaining
probabilities of the transition matrix?
5, 52 S,
Lo o 1
In canonical form this matrix can be written as
S4 5: 5, 5,
\ o \
i
_o
4
III
4
\
4
4"
2_
Thus,
1 o"
M
If the mouse is released in state 2, the expected number of times it will

occupy state 3 before reaching the absorbing state is four; for state 1, eight;
and for state 2, six.
be in state 2 is four; in state 3, four; and in state 1, eight.
be in state 2 is two; in state 1, four; and in state 3, four.

If the mouse starts in state 2, it is expected to be in 18 different states
before reaching the absorbing state; starting in state 1, 16 transitions are

expected; and starting in state 3, 10 are expected>
Practice Problem 1. Determine whether

1
2
Use the canonical forms in Exer- An absorbing Markov chain has

cise 2 to label the rows and the transition matrix
columns in terms of the original
states.
In Exercise 2, find the fundamen-

tal matrix for each absorbing
Markov chain.
For each part of Exercise 1 , find

the expected number of transi-
tions before each nonabsorbing
state reaches an absorbing state
for the absorbing Markov chains.
Use Exercise 4 to find the ex-

pected number of times that a
system starting in one nonabsorb-
ing state will be in another nonab-
sorbing state before it goes to an
absorbing state.
Introduction to Game Theory 369
Why is this an absorbing Mar- Room 4 is a trap door, so that

kov chain? What is the long- when the mouse goes to room 4,
term result of such cross-fertil- it cannot return.
ization? (a) Form a transition matrix for
this model.
11. Psychology. A mouse moves
(b) Does this transition matrix
randomly in the following maze:
represent an absorbing Mar-
kov chain?
(c) If the mouse is placed in
room 1, how many changes
X of rooms are expected be-
fore the mouse is trapped in
A mouse in room 1 has the prob- room 4?
p
abilities = 0, p= , and p = i (d) Repeat part (c) for room 2.
of moving to the various rooms. (e) Repeat part (c) for room 3.
Answer to Practice > 1. No. [0 1 0] represents an absorbing state, but it is impossible to go

Problem from state 1 to state 2 or from state 3 to state 2.
6.r Introduction to Game Theory

Overview On a third-down play with 5 yards to go, the defensive football captain says, "I believe the
quarterback will pass. Let's use our pass defense and rush." The offensive quarterback tries to
guess what the defense is going to do. He says, "They're probably thinking I'll pass. I'll call the
draw play up the middle." Notice that each opponent is trying to anticipate what the other will
do and then act accordingly. This type of strategy is present in most games, but it also occurs in
many daily activities and decisions. John Von Newmann, in the 1920s, was one of the first
people to research the theory of games. Since the theory of games and linear programming are
closely related, it was not until World War II, when linear programming was developed, that the
theory of games was developed sufficiently to be recognized as a branch of mathematics. In this
section we study

Zero-sum games

Optimum strategies
game
Value of a
Expectation
Strictly determined games
In the preceding chapters we considered decision techniques such as

linear programmingwhich we assumed a benign environment. That is, we
in
have not considered problems in which a decision-maker faces active com-
petition as she or he makes decisions. In this chapter the decision-maker
faces an active opponent a hostile, aggressive player who is trying to win.
This type of problem is similar to many of our actual experiences. Recent
development in mathematical modeling has brought many economic, politi-
cal, and social games within the scope of present-day game theory.
To understand how game theory may be useful in helping to make

decisions, consider the following example.
EXAMPLE 39 The board of education of a small city is planning to build a new school in
either location I or location II. The county board of education decides to
counter thisby building a new school in either location III or location IV.
Neither system wishes to lose students, but a careful study indicates the
following outcomes.
If the city builds in location I and the county builds in location III, the
city will gain 600 students from the county. If the city builds in location I and
the county builds in location IV, the city will gain 800 students from the
county. If the city builds in location II and the county builds in location III,
the city will lose 400 students to the county. If the city builds in location II
and the county builds in location IV, the city will gain 900 students from the
county. Where should the boards build their schools in order to be assured of
gaining the largest number or losing the smallest number of students?
SOLUTION This information may be written as the following matrix
County Builds
III IV
I 600 800
City Builds
-400 900
where the entries represent the net change in students from the viewpoint of
the city. Note from the first row of the matrix that the city will always gain
students from the county if it builds in location I. Hence, the city should
build in location I. Knowing this, the only choice left for the county is to
minimize the number of students it will lose to the city. Hence, the county
should build in location III>
The preceding example illustrates the type of problem encountered in

game theory.We now introduce the notation, terminology, and theory for
playing games. We refer to the participants as players, and the amounts that
the players gain or lose as payoffs. A matrix of payoffs (sometimes called the
payoff matrix) defines a game. The rows of the payoff matrix provide the
options available to one player (called the row player), and the columns
provide the options available to a second player (called the column player).
Let's assume there are two players, X and Y. A play of the game
consists of choosing a row of matrix A at the same time that Y chooses a
X
column of matrix A. After each play, X receives from Y the amount equal to
the element in the chosen row and column if the element is positive. If the
element is negative, Y receives this amount from X. For example, in the
matrix
X
[2-13
4 -5 2
if X should select the first row andY the third column, then X would win
three units. If Y selects the second column when X selects the first row, Y
would win one unit.
Consider a two-person game defined by the matrix
Y
2 5
-1 -2
Notice that the game is biased against player Y sinceX would probably
select the first row so that she would never lose; however, since Y is in-
volved in the game, he will do his best to make his loss as small as possible.
This comparable to a situation in which a person, finding he must sustain a

is
loss for a time, attempts to minimize this loss. X selects the first row,
because she wins with this row regardless of Y's selection. To minimize his
loss, Y must select the first column.
In each of the preceding examples each player has a fixed number
of options available on each play, and on each play the gain of one player is
the loss of the other player. Such games are called two-person zero-sum
games.
Definition: Zero-Sum A game is a zero-sum game if the sum of the gains and losses of the players is always
Game equal to zero.
By the strategy of a game for X, we mean the decision to select the

rows according to a probability distribution; the strategy for Y is the selec-
tion of the columns by a probability distribution. We indicate the strategy of
X by a row matrix P = [p, p 2 Pi Pn\, where p, + p 2 + Pi + +
pn = 1. That is, X selects row 1 with a probability of p or row 2
with a x
probability of p 2 and so forth. In a like manner, the strategy of Y is defined

,
by the column matrix
where q\ + q2 + + qm =\
Strategies A strategy in which one p, and one a, are 1 and all other p, and g, are is called a pure
strategy; otherwise, it is called a mixed strategy.
Actually a pure strategy is one in which X plays the same row and Y
plays the same column repeatedly. Of course, X would like to win as much
as possible, and Y would like to keep X from winning.
Definition: Optimum The strategies used when X and Y play their best are called optimum strategies.
Strategies
By "playing their best" we mean that X plays so that she can win as
much as possible, regardless of what Y does; similarly, Y plays so that he
can keep X's winnings as small as possible, regardless of what X does.
Definition: Value of a The value won by X (or by Y) when both X and Y play their optimum strategy is called the
Game value of the game. If the value of the game is zero, the game is said to be fair.
Consider the matrix
A =
21 #22
as defining a game in which
P = [Pt Pi] and Q

<?2
represent the strategies of X and Y, respectively. The probability that a l2

will be the payoff is p\q 2 . Likewise, the probabilities for a n a 2 and a 21 are , \
,
P\ Q\ , PiQ\ , and p 2 q 2 respectively.

.
Definition: Expected The expected value, or expectation, of the 2 x 2 matrix game is:
Value of a Game
Expectation = pi an qh + Pia-\ 2 q 2 + p 2 a2iqi + p 2 a 2 2Q2
This definition can be written in matrix form as
[Expectation] = [p\ p2] '
[21 22
In general we have the following definition of expectation:
Definition: Expectation of [Expectation] = PAQ

a Game
EXAMPLE 40 Suppose that X decides to use the strategy [\ \ ?] and Y decides to use the
strategy
Find the expectation of the matrix game
-1 3 2
4 -3 -2
1 -1
SOLUTION The expectation of the game is defined to be PAQ.

-1 3 2
[Expectation] = [i 2 ?J 4 -3 -2
1 -1
= 2-1
Expectation =
If X were to use the strategy [4 2 ?] repeatedly and Y were to use
repeatedly, the average winnings of Y would be 6, and X would lose on the

average ~6>
EXAMPLE 41 For the game

2 5
A= -1 -2
the strategies P = [1 0] and
Q =
are optimum.
2
[Expectation] = PAQ = [1 0] \
_^
= [2 5]
Expectation = 2 = value of game

The strategies P = [2 2] and
are not optimum.
2 5
[Expectation] = PAQ = [5 |J
-1 -2
2 # value of game
EXAMPLE 43 Determine whether the following matrix has a saddle point. If it does, find
the value of the game, and determine whether X or Y wins the game.
10 1 ^T~
2 1 i
-4^ 7
6 5
SOLUTION Again we circle the elements that are minimumin each row and
look for circled numbers that are maximum columns. Note that there
in their
are two saddle points for this game, but their values are the same. The value
of the game is 3, and X wins the game by selecting P = [0 1]. Y selects
1
"0
Applications
(Business and Competition for Profits. Two Competition. In a separate study
Economics) companies are deciding where to by a consultant for the competi-
locate stores in eastern Tennes- tion in Exercise 8, the following
see. If company A locates in city facts are given. If both locate in
Iand company B in city II, then city I, A receives 20% more of the
company A can expect an annual business than B. If both locate in
profit of $100,000 more than com- city II, B receives 10% more of
pany B's profit. If company A lo- the business than A. If A locates
cates in city and B in city I,
II in city I and B in city II, A re-
then B's profits will exceed A's ceives 5% more business than B.
by $50,000. If both locate in city If A locates in city II and B in city
I, their profits will be equal. If I, B receives 10% more business
both locate in II, then A's profits than A. Write a matrix for this
will exceed B's by $20,000. What competition. Is the game strictly
are the optimum strategies (from determined? If so, give the value
the standpoint of competition) of the game and strategies.
and what is the value of the
game?
and Life Sciences) 10. Political Campaign. In the race
for Commissioner of Jefferson
County, Democratic and Repub-
lican candidates can either em-
phasize their campaign in the
City of Birmingham, in the ur-
ban region around Birmingham,
or in the rural region of the
county. The units assigned to
each are gains or losses in units
of 10,000 votes.
Answers to Practice > 1. The expectation is -5.

Problems 2. The game is strictly determined; A plays the second row and B the
second column; the value of the game is 2.
6.8 Solving 2x2 Matrix Games

Overview We have already discovered that the best strategy for two-person strictly determined games is
found in the row and column that contain the saddle point. For games that are not strictly
determined, a player may choose strategies according to given probabilities. If the size of the
matrix games is 2 x 2, the probabilities are easily computed. In fact, in this section there is a
well-developed mathematical modeling theory for finding the optimum strategies for the 2 x 2
matrix game. In this section we will study
Optimum strategy for 2 x 2 games

Value of a game
Dominated rows and columns
Consider a 2 x 2 matrix game. If the game is strictly determined, the

solution can be obtained from the saddle-point theory of the preceding sec-
tion. If the game is not strictly determined, how can you find the optimum
strategy for X, the optimum strategy for Y, and the value of the game?
EXAMPLE 44 Consider the nonstrictly determined game
-3
6
Find the optimum strategies of player X and player Y, and find the value of
the game.
SOLUTION If player X chooses row with a probability p, then row 2 is

1
chosen with a probability of - p. If player Y chooses column

1 then player 1 ,
X expects to receive
E (Expectation) = 3p + 6(1 p) = 6 9p
If player Y chooses column 2, then player X expects to receive
E= 4p + (-2)0 - p) = 6p - 2
We graph these expectations with E on the vertical axis and p on the hori-
zontal. Player X wants to maximize expectations. This maximum occurs
where the two lines intersect. For any other value of p, one of the two
expectations is less. Solving these expectations simultaneously gives
Solving 2 2 Matrix Games 379
FIGURE 14
6 - 9p = bp - 2
P =
Y5
Thus, the optimum strategy for player X is
Similarly, suppose player Y chooses column 1 with a probability of q

and column 2 with a probability of 1
<y. If player X chooses row player Y
1 ,
expects
= -3<? + 4(l - q) = 4-7q

If player X chooses row 2,
E= 6q + (-2X1 - q) = Sq - 2
Setting 4 - Iq = Sq - 2 gives
</
as seen in Figure 15.

Note that E for p and E for q are identical. In fact, this value of E is the
value of the game. The game is not fair but in favor of player X, who should
use the strategy |i% i^| to win, V = f. To minimize X's expectations,
player Y uses the strategy
= 8<7-2
FIGURE 75 E = A-lq
Note that
PAQ [-
Solving 2 2 Matrix Games 381
- d-b \
+ d- b - c) Pi
(fl
a + d a + d b cj
ad be
+ = V
a + d- b - c
X wants to make wants to

the expectation as large as possible, and Y
make the expectation as small as possible. By making
term zero, the first
both players will be doing their best regardless of what the other does. X can
make the first term zero by making
Pi
a + d - b
Y can make the first term zero by making
d-b
'/i
a + d
For these strategies
ad be
a + d - b
P\
a + d- b - c
then
a- b
Pi
a + d b c
d - c
Pz = 1 - Pi
a + d a + d
Optimum Strategies For a 2 x 2 matrix game

where
-3
There are times when games defined by larger matrices may be reduced
to 2 x 2 games.
Definition: Dominated Sometimes each element of a row of a matrix is as large as or larger than the correspond-
Row ing elements of another row. When this happens, the row with the larger elements is said
to dominate the other row, and the row with the smaller elements is said to be dominated
by the other row.
Of course, X would rather play the row with the larger elements be-
cause X would like to win as much as possible. Thus, the dominated row
may be omitted without affecting the solution of the matrix game. Similarly,
we say one column dominates another if the elements of the dominating
column are as small as or are smaller than the corresponding elements of the
dominated column. Since Y would always choose the column containing the
smaller elements, the dominated column may be omitted from the game. The
dominated rows and columns may be dropped in any order.
EXAMPLE 48 For the matrix game
1 4 -3
-1 -3 -4
-3 7 5.
each element of the first row, [1 4 - -3], is larger than the corresponding
element of the second row, [-1 -3 -4]; hence, the second row may be
omitted to obtain
1 4
-3 7
Each element of the first column
is less than the corresponding element of the second column
4
7
so the second column may be omitted, and we have

1 -3
-3 5
The game is biased in favor of Y since the value of the game is
1(5) - (-3K-3) 1
1 + 5 - (-3) - (-3)
Solving 2x2 Matrix Games 385
For the reduced 2x2 game, X's strategy should be [ jj, and Y's strategy
should be
Relative to the original matrix, X's strategy should be [ gj, and Y's
strategy should be
The zero probability indicates that X should never play the second row and
Y should never play the second column.
Practice Problems > 1. Find the optimum strategy for X and Y and the value of the game defined
by
5 2
1 3
2. For the game defined by
12 7
6 5
-4 4
find the optimum strategies and the value of the game.

Exercise Set 6.8 1. Find the value of each matrix 4 -12

(f)
game and optimum strategy
the -2 9
for each player. Be sure to
For the following matrix games,
check whether the game is
find the optimum strategies and
strictly determined.
the values of the games.
(a) 16 8
[l-l]
(a) 7 6
2
(b)
-3 3
6 3 -3 -3
4 (b) 2 2 2
5 4 -5 8
(d)
12 -3 8 9-5
-2 2 -3 5 6
(O
5 6 7
(e)
[5 1] 10 10 -3
Solving 2x2 Matrix Games 387
10. Advertising. Two furniture

stores compete for sales in a
small town through radio and
newspaper advertising. A mar-
keting consultant has indicated
the percentage gain or loss in
sales when the weekly advertis-
ing of each specializes in a given
medium.
6-9 A Technique for Solving x 2 Games
Overview In the preceding sections we discussed the solution of 2 x m or n x 2 games when the games
were strictly determined, or could be reduced to 2 x 2 games by eliminating the dominated rows
and columns. We now utilize graphical means to solve 2 x m and n x 2 games that are not
strictly determined and cannot be reduced to 2 x 2 games by dropping the dominated rows and
columns.
In an n x 2 matrix game, player X has n strategies and player Y has two

strategies. In a2 x m game, player X has two strategies and player Y has m
strategies. Both types of games can be solved graphically.
EXAMPLE 49 Find the optimum strategy for X, the optimum strategy for Y, and the value
game by graphical means.
of the following matrix
Y
3 -2
-2 3
solution Recall that, if P is the optimum strategy for X and Q is the opti-
mum strategy for Y, then PAQ = [V], the value of the game. If Y should
play a strategy Q', which is not optimum, and X continues to play an
optimum X's winnings will be greater than or equal to V; that

strategy, then
is, PAQ' = [V] where V
s V. Now denote by [p 1 - p] the optimum
strategy for X. This is a suitable representation for the probability matrix
because the sum of the probabilities must equal 1, andp + (1 - p) = 1. If Y
plays the first column, then the expectation, or expected winnings, for X
would be
[p 1 -,][. = [V]
[P 1-P] _ [V]
3/7-
A Technique for Solving n x 2 Games 389
FIGURE 16
value of V occurs at the intersection of
V= 1 - 2p and V = 5p - 2
p = f and V = yat>this point. Thus, the optimum strategy for X is [f fj. and
the value of the game is 7-
To find Y, note that the optimum strategy for
the optimum strategy for
X was determined by computed from the first and
the intersection of lines
third columns of the matrix. Omit the second column and consider the 2 x 2
matrix
3 -1
-2 1
Y's strategy is computed to be
1- (-1) 3 - (-2)
= and qi
=
<7i
3 + 1 - (-2) - (-1) 7 3 + 1 - (-2) - (-1)
Thus, Y's optimal strategy is
EXAMPLE 50 Find the optimum strategy for X and Y and the value of the following game:
3
4
-1
-3
SOLUTION Let Y's optimum strategy be
q
1 -q
Then the expectation of Y will be less than or equal to V, and the expectation
will be
[ 3
A Technique for Solving n > 2 Games 391
This point is q = i, V= |. Thus, the optimum strategy for Y is
and the expected earnings for X are f To find the optimum strategy for X,
.
note that the optimum strategy for Y was determined by the intersection of
lines computed from the first and fourth rows of the matrix. Omit the second
and third rows and consider the 2 x 2 matrix
3 1
-3 3
Then
3 - (-3)
P\
= and P*
3 + 3 - (-3) - 1
Thus, the optimum strategy for X is [| i]>
Exercise Set 6.9 For the following matrix games,

find the optimum strategies for
X and Y, and the values of the
games. Be sure to check for sad-
dle points.
5
(a) -3 4
-5_
i 3
(b)
10 I 2
(O
(d)
(e)
If)
Solve the following two-person ma-

trix games with the given payoffs:
\l
Solving Games by Linear Programming 393
Determine the relative percent- the three antibiotics for the

age of land the farmer should de- greatest probability of killing
vote to each crop. bacteria?
10. Medicine. Suppose there are

two bacteria associated with a
disease. Three antibiotics have
been tested on each bacterium
and the probability of destroying
the bacterium is given. What
should the strategy be for mixing
where V is the value of the game. Since q\ + qi + + qm = 1.
x, + x2 + + xm =
v
The objective function to be maximized is
-=/=*,+
1
x2 + + x
The constraint inequalities are
l|.Vi + Cl\ 2 X2 + + a Xm x m < 1
a n \X\ + a2-"'2 + " ' + a m x, ^ 1
x, >
xm >
where ay is an element of the matrix game.
EXAMPLE 51 Use the linear programming techniques to find strategies for the matrix game
-3 4
6 -2
SOLUTION First note that not all the elements in the payoff matrix are posi-
tive. To make all the elements positive, add 4 to each element.
10
The constraint equations are
Xi + 8.v 2 < 1
lO.v, + 2.v 2 < 1
Xi > .v 2 >
We desire to maximize
f=X] + x2
To apply the simplex method of linear programming, we introduce the

slack variables y\ and y2 (each >0) such that
X] + 8x2 + V| = 1
10a, + 2.v 2 + y: = 1
Thus, the first tableau becomes
X\ X2 Vi V; /
1 8 10 1'
10) 2 10 11
l-i"~TT 'o~iT o
The second tableau is

78
= 1_
yi
78
Thus,
P\ = y\V =
Pi = y2V
Player X thus has the strategy [-ft ^J, player Y has the strategy
and the value of the game is
78
1?
EXAMPLE 52 Use linear programming techniques to find the strategies for the matrix game
2
1
6*, + 2*2 + 5* 3 + y, = 1
3*1 + 5*2 + 3*3 + v2 = 1
.v, + 8*2 + 4*3 + y3 = 1
Thus, the first tableau becomes
*\
The strategy of player Y is thus
From the dual problem,

(f)
(g)
(h)
Advertisement. Two competitive

Direct
automobile dealerships in a small
Radio Paper mail
town compete for the town's
Radio 2
business through radio, newspa-
Paper -3 -1
per, and direct mail. A marketing !1
consultant has indicated the per- Find the optimum strategies for A
centage gain or loss in a sales and B and the value of game.
for each weekly advertisement
choice. Dealership B does not use
direct mail.
Applications
(Social and 7. Agriculture. A farmer in Georgia
Life Sciences) grows peanuts and cotton. The
following matrix gives gross in-
come in thousands of dollars un-
der different conditions.
of Moody's, Value Line, and other investment services reveals that 60%
believe the market will increase, and 40% believe the market will decrease in
the next year.The table gives the expected payoff (sometimes called ex-
pected monetary value) of buying each of the three stocks.
PAYOFF TABLE
Success in solving application problems with probability depends on

the ability to use the correct formula. Review the conditions that allow you
to use each of the following formulas:
Important P(A n B) = P(A,)

P(B 2 \a,) + P(B^) p(a 2 \b,)
Formulas P(Bi) P(A\B,)
P(B,\A)
P(B0 P(A\B\) + P(B 2 ) P(A\B 2 ) + + P(B n ) P(A\B)
E(x) = X\p(x\) + x2 p(x 2 + )

+ xp(x n )
= d - c
p{x) = C(n, x)p x q n *
P\
d - b =
P2
'/i
a + d - b - c
a c
V=
a + d - b - c
random, and a chip is drawn.

Suppose we know only that a
red chip was drawn. What is the
probability that it was drawn
from box A?
7. Find the steady-state row matrix

for the Markov process whose
transition matrix is as follows:
(a)
able denoting the sum of num- 18. A shipment contains 96 good

bers on two chips drawn at items and 4 defective ones.
random, tabulate the probability Three items are drawn from the
function; find E(x). shipment. What is the proba-
bility
15. Box 1 contains 10 chips with 5
(a) That all 3 are defective?
marked A, 4 marked B, and 1
(b) Of exactly 2 defective
marked C. Box 2 contains 4
items?
chips: 2 A's, B, and C. Urn
1 1 A
(c) Of exactly defective item?
1
contains 3 red balls and 2 black
(d) Of no defective items?
balls, urn B contains 2 red and 2
black balls, and urn C contains 1 19. According to a mortality table,
red and 4 black balls. A coin is the probability John Sails will
tossed. If the coin shows heads, live for 1 year is .992. He buys a
a chip drawn from box
is if it 1 ; $10,000 1-year term life insur-
shows tails, a chip is drawn from ance policy for $40. What is the
box 2. When a chip is drawn expected gain or loss of the in-
from either box, a ball is drawn surance company? (Assume no
from the urn denoted on the interest rate.)
chip. The experiment is per-
20. Find the optimum strategies for
formed. What is the probability
X and Y
and the value of the
of getting a red ball?
game defined by
16. Consider the experiment in Ex-
5
ercise 15 with the following
-3
changes. If a black ball is drawn
8
from urn A, it is placed in urn B,
and a second ball is drawn from 21. The movement of a mouse in a
urn B. What is the probability of compartment maze is defined by
getting a red ball? the transition matrix
17. Consider the experiment in Ex-

ercise 15 with the following
changes. If a black ball is drawn
from urn B (at the time the chip
indicates drawing from urn B), it
is placed in urn A and a second
ball is drawn from urn A. What

is the probability of getting a red
ball?
Chapter Test 405
Chapter Test
1. Consider the following tree diagram. Find P(A\D).
Start
2. Three machines. A, B, and C, produce 50%, 30%, and 20%, respec-

tively, of all the parts in a small motor. It has been found that 3% of the
items produced by A, 2% of the items produced by B, and 1% of the
items produced by C are defective. What is the probability that some
part of a small motor is defective?
3. Find the steady-state probability matrix for
4. Eight percent of all residents of Orange County have the flu virus. The
probability that a medical examination will indicate the disease if a per-
son has the virus is .8. and the probability that the examination will
incorrectly indicate the disease is .3. What is the probability that Sue
actually has the flu if her examination indicates that she is sick with the
flu?
5. Discuss whether or not

8. A lot of 12 tires contains 3 defective and 9 good tires. What is the

probability that 3 good tires will be selected at random?
9. Rework Exercise 8 for the probability that 1 good and 2 defective tires
will be selected at random.
10. Find the value of the game
11. Find the optimum strategies for the game given in Exercise 10.
12. Find the value of the game

^32 5'
-1 1 -3
.53 4
13. Find the optimum strategies for the game given in Exercise 12.
14. Discuss whether or not

Statistics
When Aunt Jane asserts that smoking is not harmful to one's health because
Uncle Joe lived to be 88 years of age and smoked two packs of cigarettes
every day of his adult life. Aunt Jane is using a probability or statistical idea.
That is. Aunt Jane organized the data of her experience (Uncle Joe) and then
made a statement on the basis of her data. However, she lacked understand-
ing of what data were needed, how the data should be organized, and what
conclusions were appropriate or inappropriate relative to the data. This
chapter should help you avoid making the same types of mistakes that Aunt
Jane made.
Statistics can be divided into two subdivisions: descriptive and inferen-
tial statistics.
Descriptive Statistics Inferential Statistics
Techniques are used to summarize and Generalizations or conclusions are made

describe the characteristics of a set of about the data in a large group (called
data. the population) from a small portion

(called a sample).
we will study descriptive

In the first three sections of this chapter,
understanding a set of data by forming frequency distributions,
statistics:
drawing associated graphs, finding measures of central tendency (mean,
median, and mode), and finding measures of the scattering of data.
407
408 Chapter 7 Statistics
Inferential statistics is introduced by studying the normal distribution,

which is the foundation of much of the research studies in business, eco-
nomics, and the social and life sciences. Two sections introduce some of the
theory of using the normal distribution in inferential statistics.
In this chapter, we consider such diverse applications as marketing,
pollution, psychology, politics, quality control, and drug effectiveness.
7.1 Frequency Distributions and Graphical Representations
Overview We are immersed daily in a torrent of numbers flowing in bubbling splendor from our televi-
sions, radios, newspapers, hair stylist, and our favorite Uncle Al. Although data are a part of our
daily life-style, it is evident that we often do not know how to organize, interpret, or understand
the message being conveyed. In this section, we learn to organize and summarize data for better
understanding of statistics.
First, we introduce some basic techniques for classifying and summarizing a set of
observed measurements. Then we will represent the data with graphs. In fact, we will construct:
Tables representing frequency distributions

Relative frequency distributions
Bar graphs
Line graphs

Histograms
Frequency polygons
Circle graphs
The first objective of a statistician is to develop a plan to collect data

for a study. After the data are collected, the second objective of a statistician
is to this large amount of data. Suppose you have col-
make sense out of
lected the Table 1 The table contains your tabulation of the
numbers in .
number of colds experienced during a winter by each of a group of 30

elementary school children.
711034553233662
TABLE 1
421003456314134
A quick glance at this array of numbers tells us very little about what
the data imply about the group of people represented. Closer observation
indicates that the largest number of colds experienced was 7, and the small-
est number was 0. The difference between the largest and smallest entry in
the data is called the range of the data. In this case, the range is 7 - = 7. To
understand better the significance of this list of numbers, we might summa-
rize it in a frequency distribution. To make this summary, we record the
number of students who reported each number of colds.
Frequency Distributions and Graphical Representations 409
EXAMPLE 1 Make a frequency distribution for the data in Table 1.
SOLUTION From the summary in Table 2 it is easily seen that three colds was
the number most often reported by a child. The summary also shows how
the number of colds was distributed among the 30 students.
TABLE 2
Number of Colds
We arbitrarily select six classes for our grouping. Since 17 * 6 is 2.833, the
length of the classes (if the classes are of equal length) must be more than
2.833 in order to include all the data in six classes.Whenever feasible, it is
desirable to have classes of equal integral length. Thus, we arbitrarily select
the following class limits: 1-3, 4-6, 7-9, 10-12, and so on. The grouped
frequency distribution is found in Table 4.
TABLE 4
Class
SOLUTION
Interval
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986
Year
FIGURE 1
A line graph shows the fluctuations and emphasizes the changes that
have taken place better than a bar graph. The line graph in Figure 2 repre-
sents the distance (in meters) run in 6 minutes by a group of freshmen in a
physical education class. Looking at this graph, we can readily see the
variation in the number who ran given distances in 6 minutes.
745 845 945 1045 1145 1245 1345 1445 1545 1645
FIGURE 2 Number of meters
Sometimes, data are so erratic that a line graph would have little mean-
ing. In this case, a scattergram improves the understanding of the data. A
scattergram consists of data points and a trend line drawn to visually "fit"
the data. Figure 3 shows data relating ACT scores in mathematics and stu-
dents' averages in a class at the end of a semester. A trend line is drawn with
about the same number of points on each side of the line and with points
scattered on each side of the line at both ends of the line. Different lines can
be drawn visually, but usually most of these lines will have approximately
the same slope. The scattering of the points about a trend line shows how the
data correlate. From Figure 3, there is a positive correlation: as the ACT
mathematics score increases, there seems to be a corresponding increase in
the class average of a student. If the slope of the line were negative, there
would be a negative correlation. For perfect correlation, all points would be
on the same line; that is, there would be no scattering about the line. When
the points are so scattered that it is nearly impossible to draw a trend line,
then we say that there is very little correlation.
100-
SOLUTION Figures 4(a) and (b) are identical representations of a histogram

except that in (a) each rectangle is denoted by the class interval and in (b)
each rectangle is denoted by the middle of the class interval, called the class
mark>
7-
One of the simplest types of graphs is the circle graph, sometimes

called a pie chart. It consists of a circle partitioned into sections, where each
section represents a percentage of the whole.
EXAMPLE 7 Table 9 records the final-examination grades in a class. Represent the data
by a circle graph.
TABLE 9
Final-Examination Grade
Practice Problems > 1. For the data in Table 8, draw a histogram on a graphing calculator. Use X
Min: 15, Max: 40, Scl: 5, Y Min: 0, Max: 8, Scl: 2. How does your graph
compare with Figure 4?
2. To draw a ne graph on top of a hi stogram using a graphing calculator,
l i
use Graph SHIFT Line EXE

1
1
1 Draw a line graph for Table 8.
1
1
[
|
|
.
3. For the grouped frequency distribution given in Table 6, draw a histo-

gram.
Answers to practice problems are found on puge 421
Exercise Set 7.1 1. In a transportation survey, bus (d) Find the number of mothers
riders on the Friday evening run whose ages were between 19
were asked how many times and 34, including the end-
they had ridden the bus that points.
week. Summarize the data in a (e) Find class marks.

frequency distribution. 3. In the given pie chart or circle
graph
(a) What percentage are profes-

sionals?
(b) What percentage are crafts-
men?
3
(c) What percentage are man-
agers or clerical?
The following is a tabulation of (d) What percentage are neither
the ages of mothers of the first managers nor professionals?
babies born in Morningside Hos-
pital in 1990:
Class Tally Ftequency
15-19 mi
20-24
25-29
M
M
30-34 1
35-39 II
(a) Can you determine the num-

ber of mothers in the tabula-
tion?
(b) Find the number of mothers
younger than 30.
(c) Find the number of mothers
who were at least 20 years of
age.
4. The following table gives the

number of students who had a
specific number of absences in a
given semester.
Number of

the age of a student at
Roswell College is between

20 and 24?
the age of a student is less
than 35?
the age of a student is more
than 19?
(d) What is the probability that
the age of a student is less
than 40?
(e) What is the probability that
the age of a student is
greater than 14?
Park officials want to under-

stand the use of a municipal
park. One evening the officials
interviewed 36 people and re-
corded their ages.
(a) Summarize the resulting

data in a grouped frequency
distribution with 7 intervals
of equal length. Let the first
interval be 4-14.
(b) What trend can you see in
the data?
7
school seniors on a mathematics

placement test. Tabulate this in-
formation into five groups of
minimum integral length, with
the first class beginning at 450.
477
Applications
(Business and
Economics)
15. In 1990, how many more model

Acars were sold than model B?
16. In which 4-year period did
model C have the greatest de-
crease in sales? What was the
decrease?
17. In which 4-year period did
model B have the greatest in-
crease in sales? What was the in-
crease?
18. Compare the increase in sales of
the three models from 1974 to
1990.
19. Sales. Tabulate a relative fre-
quency distribution from the ta-

ble, where the variable is the
number of units sold per day.
Units Sold
(c) What is the probability of a (d) What is the probability of a

price-earnings ratio of less price-earnings ratio be-
than 35? tween 1 5 and 29, inclusive?
and Life Sciences) 23. Classification of Occupations.
Use a protractor to construct a
pie chart showing the percent-
age of women who work in the
various occupations. These per-
cents are given below.
Professional
2. Did you get the graph in Figure 5?
3. /
7--
4--
2
-
13-19 20-26 27-33 34-40 41-47
7.2 Measures of Central Tendency
Overview In the preceding section, we used histograms, frequency polygons, and pie charts to summarize
and explain sets of data. Sometimes we need a more concise procedure for characterizing a set
of data. In fact, we want a simple number that estimates the location of the center of a set of
data. Therefore, in this section, we introduce averages, or measures of central tendency.

Sandra receives 69, 71 , 78, 82, and 73 on her five tests in Math 1 02. She gives her average
grade as 74.6, but her friend Sam claims her average is 73. Which average is correct? In this
section, we learn that both averages are correct. Sandra found the mean and Sam the median.
One common use of statistics is comparing sets of data with averages (or, more accu-
rately, measures of central tendency). Three measures are in general use the arithmetic
mean, the median, and the mode. The fact that there are these three (as well as others) often
leads to misuses of statistics. One measure may be quoted, and the reader automatically thinks
of another. When a measure is quoted, immediately ask the question, "Which one?" In this
section, we will study
The arithmetic mean

The arithmetic mean for a frequency distribution
The median
The mode
The most widely used measure of central tendency is the arithmetic

mean (sometimes called arithmetic average). The arithmetic mean of a set of
n measurements is the sum of the measurements divided by n.
Definition: Consider n measurements x-i ,

x2 ,
x3 xn . The formula for the arithmetic mean,
Arithmetic Mean denoted by x, is
. x, + x2 + x3 +
+ x
Measures of Central Tendency 423
This formula is sometimes written in sigma notation as
x
_
= t* or
2>
n n
That is,
2 -V = X
/=]
Xi = x, + x2 + + x
EXAMPLE 8 Find the arithmetic mean of 8. 16, 4, 12, and 10.
SOLUTION _ 8+16 + 4+12+10 ,

=10
x = -
5 ^
EXAMPLE 9 Find the arithmetic mean of 25, 25. 25, 25, 30, 30, 30, 40. 40, 40, 40, 50.
SOLUTION
__ 25 + 25 + 25 + 25 + 30 + 30 + 30 + 40 + 40 + 40 + 40 + 50
x -
j2
25(4) + 30(3) + 40(4) + 50(1)

=
4+3+4+1
- fOO =
"12 3
In Example 9, observe that the 4, 3, 4, and 1 are the frequencies of the

25, 30, 40, and 50, respectively. is obtained by multiply-
Note that the mean
ing each value by the frequency of occurrence of the value and by dividing
the sum of these products by the sum of the frequencies. Let us now general-
ize the formula for finding the arithmetic mean to include frequencies of the
observations.
Definition: Arithmetic Let x u x 2 x m be different measurements.
Mean, Frequency
. *ifi +x 2 f2 + x 3 f3 +
+ x m f
Distribution
U + f2 + + f
where f, is the frequency of x, for / = 1, 2, 3, . . ., m.
Sometimes this formula is written as
x = ~^ or
"v7
CALCULATOR NOTE Most calculators have a statisticalmode that makes it easy to compute
arithmetic means. In the Appendix we will discuss the statistical mode for a
Casio calculator.
If data are presented in a frequency table, we may have no way of

knowing the distribution of the data within a class. We therefore assume that
the data are uniformly distributed within a class interval around the class
mark or that all of the data within a class interval are located at the class
mark. Thus, we use the formula for x given above, where x represents the
class mark, / the frequency of each class, and m the number of class in-
tervals.
EXAMPLE 10 Find the arithmetic mean for the data in Table 10.
SOLUTION TABLE 10
Class Mark (x)

EXAMPLE 11 Consider the set of five measurements 7, 1,2, 1,3. Arranged in increasing
order, they may be written as
1, 1,2, 3,7
Median
Hence the median is 2>
EXAMPLE 12 The array
25,2, 5,6, 5, 23, 7, 10,22, 15,21,23
can be arranged in decreasing order as
25, 23, 23, 22, 21, 15, 10, 7, 6, 5, 5, 2
Median
So the median is
15 + 10
= 12.5
The third measure of central tendency is called the mode, the measure-
ment that appears most often in a given set of data.
Definition: Mode The mode of a set of measurements is the observation that occurs most often. If every
measurement occurs only once, then there is no mode. If two measurements occur with
the same largest frequency, the set of data is bimodal. It may be the case that there are
three or more modes.
EXAMPLE 13 Baseball caps with the following head sizes were sold in a week by the Glo-
Slo Sporting Goods Store: 7, 7$, 8, 6, 7i, 7, 6i 8i 1\, 8, 1\. Find the mode
head size.
SOLUTION The mode is 1\ it occurs 4 times, more times than any other
size>
The decision about which measure of central tendency to use in a given

situation is not always easy. The mean is a good average of magnitudes, such
as weights, test scores, and prices, provided there are no extreme values
that distort the data. When extraordinarily large or small values are included
in the data median is usually better than the mean. However, the
set, the
mean is the average most often used, since it gives equal weight to the value
of each measurement. The median is a positional average. It replaces rank-
ing numbers. The mode is used when the "most common" measurement is
desired. The most appropriate measure for the price of pizzas in town would
be the arithmetic mean. However, to select the best-tasting pizza in town,

one could use the mode if a large number of persons were involved, or one
could use the idea of the median classifying one-half of the pizzas as tast-
ing better and one-half not so good. Unfortunately, people use the average
that is suitable to the objectives they hope to accomplish. A factory might
report an average income of $25,000, whereas the union representative
might report only $22,000. Don't be misled. Determine whether the statistic
is the mean, the median, or the mode.
EXAMPLE 14 In one series of games against the Dodgers, the Reds won 6 of 7 games by the
following scores (see Table Find the mean, median, and mode of these
11).
scores. Discuss which is the better measurement of central tendency.
TABLE 11
Dodgers I 2 6 1 15 4 2 2
Reds 14 7 2 1 5 3 3
solution Mean score of Dodgers =

4.57; mean score of Reds = 3.57; me-
dian score of Dodgers =
median score of Reds = 3. The mode scores are
2;
the same as the medians. The median score is more usable than the mean
score.
The formula for the mean of a frequency distribution assists in under-

standing the mean of a probability distribution. Suppose we write
- _ /i-v, + f2 x2 + + fm xm
fx +h +
The average salary of 6 office workers in the XYZ Corporation is $14,000.

John remembers 5 of the 6 salaries: $10,000, $13,000, $19,000, $16,000,
and $12,000. If the average is the median, calculate for John the missing
salary. Do the same if the average is the mean.
Exercise Set 7.2 1. Compute the arithmetic mean, (c) The mean, median, and
the median, and the mode (or mode are equal.
modes if any) for the given sets (d) The mean and median have
of data. values of 8.
(e) The mean and mode have

(a) 3, 4, 5. 8, 10
values of 6.
(b) 4. 6, 6, 8, 9, 12
(f) The mean, median, and
(c) 3, 6, 2, 6, 5, 6. 4, 1, 1
mode have values of 10.
(d) 7, 1, 3, 1,4,6, 5, 2
(e) 21, 13, 12,6, 23, 23, 20, 19 5. The mean score of a set of 8
(f) 18, 13, 12, 14, 12, 11, 16, 15, scores is 65. What is the sum of
21 the 8 scores'?
6. The mean score of 9 of 10 scores

2. An elevator has a capacity of 15 is 81. The tenth score is 100.
people and a load limit of 2250 What is the mean of the 10
lb. What is the mean weight of scores?
the passengers if the elevator is
7. Which of the three averages
loaded to capacity with people
should be used for the following
and weight?
data?
(a) The average salary of 4

3. Find the mean of the given dis-
salesmen and the owner of a
tribution.
small store
(b) The average height of all
Frequency
male students in We-Fail
10 High School
20
(c) The average dress size sold
30
at Acme Apparel
40
8. At the initial meeting of an ath-
4. Make up a set of data with 4 or letic club, the weights of the
more measurements, not all of members were found to be 220,
which are equal, with each of 275. 199,246,302,333,401, 190,
the following characteristics: 286, 254, 302, 323,221.
(a) The mean and median are (a) Compute the mean, median,
equal. and mode of the data.
(b) The mean and mode are (b) Which measure is most rep-
equal. resentative of the data?
9. The weights in kilograms of the 12.

members of the Laramy High- 40--
School football squad are as fol-
30--
lows: 75, 60, 62, 94, 78, 80, 72,
20
74,76,89,95,98,97,80,98,91,
96, 90, 84, 73, 80, 92, 94, 96, 99, 10
84, 60, 68, 74, 80, 92, 96, 88, 74,

70 80 90 100 110
84, 94, 72, 76, 64, 80. Score
(a) What is the mean weight of

Find the mean or the expected value
the football squad?
of the probability distributions de-
(b) What is median weight?
the
fined by the following:
(c) What is modal weight?
the
(d) If you were a sportswriter 13.
assigned to do a story on this P(x)
squad, how would you de-

1.00-
scribe the (average) weight?
10. The given grades were recorded

for a test on this chapter. Find
the arithmetic mean.
Score Frequency
100
90
80
70
60
50
40
score from the mean, considered

Score Frequency
as signed numbers) is 0. Show
140-149
that this statement is true for the
following data: 5, 8, 10, 12, 15.
17. The table shows the distribution

of scores on a test administered
to freshmen at Laneville Col-
lege. Find the arithmetic mean.
and Life Sciences) 23. Psychology. Find the mean of Number of
the following test scores (scores
are class marks of a frequency
distribution).
Test Score Frequency
50
60
70
80
90
24. Expense Accounts. The follow-

ing data have been collected on
the expenses (excluding travel
and lodging) of 6 trips made by
teachers in the mathematics de-
partment at Snelling College.
Let
$64.30
20
$3.22 average expense

per day
$148.00
20
= $7.40 average expense

per day
_ $64.30
= $12.86 average expense

per day
Which average is realistic?

Measures of Variation 431
Answers to Practice i . Program: |

MODE |
[x] |
SHIFT |Scl| |EXE| 8 [DT] 16 [of] 4 [DTJ 12
|
[Dfl 10 [DT] |
ALPHA 1
fw| [EXE] SHIFT [7] EXE Did you get n =
|
|
|
|
.
5;x = 10?
2. jc = W= 23.5
3. Program: [MODE] p x] SHIFT Scl [EXE] 4 [pf] [pf] 14 SHIFT] p] |
1
1
|
1
8 [Df] 24 SHIFT Q] 20 fpfl 34 [SHIFT] Q] 10 [W] ALPHA [W]

|
|
|
|
jEXE| SHIFT [7] |EXE| Did you get n = 40; x = 23.5?

|
|
.
4. For the median a $15,000 salary is missing. For the mean a $14,000 salary
is missing.
7.3 Measures of Variation
Overview The fact that the mean salary of the Doran Company exceeds the mean salary of the Wargo
Company does not imply that the salaries of the Doran Company are superior to those of the
Wargo Company. Compare four monthly salaries: $8000, $8100, $8100, and $25,000, with a
mean of $12,300; and four salaries: $11,800, $11,900, $12,000, and $12,200, with a mean of
$1 1 ,975. The mean salary of the first company exceeds the mean salary of the second company,
but the lowest salary of the second company is better than all the salaries except the largest
salary of the first company.

Joe Smith had a score of 450 on an intelligence test. This score is considered exceptional
because it is well above the mean of 400. Is it? The answer is no because, on this test, 45% of the
scores are between 450 and 550. To be exceptional on this test, one must score above 500.
The average depth of the Cahaba River is 1 foot. This river should be a nice river in which
to go wading. Wait a minute! There are many shallow areas, but there are also a number of
potholes 15- to 16-feet deep.

The preceding examples indicate the need for a measure of dispersion, or scattering, of
data. That is, we need a measure to indicate whether the entries in a set of data are close to or
not close to the mean. Thus, in this section we consider as measures of scattering
The range
Variance
Standard deviation
Standard deviation for a frequency distribution
Standard deviation of a probability distribution
i scores
Chebyshev's theorem
There are several ways to measure dispersion of data. The measure of scat-
tering that is easiest to calculate is the range, introduced in Section 1.
EXAMPLE 15 For the set of data, 7, 3, 1, 15, 41, 74, and 35, the range is 73 since
74 - 1 = 73>
Although the range is easy to obtain, it is not always a good measure of

dispersion because it is so radically affected by a single extreme value. For
example, suppose the 74 in the set of observations listed previously was
miscopied and listed as 24 instead. Note that the range changes from 73 to
40.
Since the range is affected significantly by extreme values, other mea-
sures of scattering or dispersion are preferable. In this section, we consider
variance, denoted by si, and standard deviation, sx , since these measures of
scattering are used in inferential statistics.
Variance Variance for a set of data can be obtained in four steps.
(a) Compute the mean x.
(b) Compute the difference in each observation and the arithmetic mean, x - x.
- 2
(c) Square each difference, (x x) .
(d) Divide the sum of the differences squared by n - 1, where n is the number of
observations.
EXAMPLE 16 Find the variance for the data 5, 7, 1, 2, 3, and 6.
SOLUTION
(a) Compute the mean of the data. (See Table 12.)
(b) Determine the difference between each x and x = 4. (See the second
column of Table 12.)
(c) Compute the square of each of these differences; that is, compute (x -
x) 2 (the third column of Table 12).
(d) Sum the squares of differences (that is, sum the third column of Table 12)
and divide by n 1=6 1 = 5,
The four-step calculation above is equivalent to the following formula

for variance.
Formula for Variance Variance, denoted bys*, is
- 2 - 2 - 2
2
(x, x) + (x 2 x) + (x 3 x) +
- 24 124 - 6(4) 2 2jc 2 = 124,

A
6 - 1 n = 6
28
= 5.6
This is of course the same answer obtained in Example 16.
TABLE 13
X
TABLE 14
X
where fjn is interpreted as p(x,) and x is replaced by its probability distribu-

tion counterpart,
E(x) = x\p(x\) + x 2 p(x 2 + )

+ xn p(x)
However, most of the time for probability distributions, E(x) is replaced by

the Greek letter p., (mean of a probability distribution).
Definition: Variance of a If the probabilities of outcomes x,, x 2 , . . ., x are p(x^), p(x 2 ), . .
Probability Distribution variance of the probability distribution is

Definition: z Scores A score or measurement, denoted by x, from a population with mean x (or jix and
standard deviation s x (or <r x ) has a corresponding z score given by
x - x x - tl x
z = or z =
the number of standard deviations from the mean.
The c score is if .v and x are

a measurement without units. For example,
in feet, then .v,and the division eliminates the units. Consequently,
is in feet,
; scores are of value in comparing two sets of data with different units. In
many comparisons, the mean and standard deviation of the population are
not known, and the z score is approximated by using the mean and standard
deviation of a sample.
EXAMPLE 22 Teresa scores a 76 on the entrance test at school X and an 82 at school Y. At

which school did she have the best score?
SOLUTION To answer this question, we need to know that the mean score at
school X was 70 with a standard deviation of 12 and the mean score at school
Y was 76 with a standard deviation of 16. The z scores are then as follows:
76
School X:
School Y: z
|i - ka (i \l + ka
1
I 1
Mean of a sample
I
I
u .
J
FIGURE 7
EXAMPLE 23 Suppose you know the mean life of a battery is 48 months with a standard
deviation of 6 months. What percentage of the batteries produced should last
between 36 months and 60 months?
SOLUTION In Figure 8 note that

12. The mean of a population is 100

with a standard deviation of 10.
Convert the following to z

scores. 20. 5
(a) 110
(b) 80
(c) 71
(d) 120
(e) 140
(f) 40
13. Find the standard deviation of

the following sample data:
Class
(a) How many of the values statistics final, he received a 90.

from class I lie within 2 stan- where the mean grade was 80
dard deviations on either and the standard deviation was
side of the mean? What per- 15. If the standardized scores
centage of class I is this? (i.e.. scores adjusted to a mean
(b) How many of the values of and a standard deviation of
from class II lie within 2 I) were the same in each case,
standard deviations on ei- then .v
=
ther side of the mean? What
(a) 10
percentage of class II is this?
(b) 12
27. Exam. A student received a (c) 16
grade of 80 in a math final where (d) 18
the mean grade was 72 and the (e) 20
standard deviation was s. In the
Applications
(Business and 28. Quality Assessment. The fol-
Economics) lowing data show the miles per

gallon reported by owners of 5
eight-cylinder automobiles from
different manufacturers.
Manufacturer
29.
The Normal Distribution 441
Score
33
The normal distribution

Properties ot the normal curve
How to use the normal curve
Applications of the normal curve
One of the fortunate surprises in statistics is that many line graphs and
bar graphs are approximately bell-shaped. In fact, if we modify some line
graphs to indicate probability rather than frequency, the resulting graphs will
closely approximate a smooth, bell-shaped curve called the normal probabil-
ity curve. If this is true, the data involved are said to be normally distributed.
We now plot a bar graph and a line graph for the data given in Table 15
using probability rather than frequency. (See Figure 9.) The height of each
rectangle represents the probability that the variable falls in that interval.
For example, the probability that the variable falls between 19.5 and 24.5 is
.30. However, for a probability curve, we want the area of the rectangle
instead of the height to represent the probability. Since the width of each
rectangle is 5, the area of each rectangle is 5 times the probability. We make
the area of each rectangle equal to the probability by dividing the height of
each rectangle by 5. The resulting graph is drawn in Figure 10. This smooth
curve is an example of a normal curve. Note some interesting properties of
the normal probability curve by studying the approximating bar graph in
Figure 10.
TABLE 75
Class Frequency Relative Frequency
5-9
FIGURE 10
The sum of the areas of all the rectangles is I. Therefore, it is reason-

assume that the area under the smooth, approximating curve is
able to The I .
mean of the data is 22, symmetric about a vertical line

and the curve is
through the mean. The standard deviation of the data is approximately 7.5.
The interval about the mean that extends for one standard deviation on
either side of the mean is the interval from (22 - 7.5) to (22 + 7.5). This
interval from 14.5 to 29.5 contains 20 + 20 + 20 = 20, or 70%, of the data (see
Figure 10). The interval about the mean that extends for 2 standard devia-
tions on either side of the mean is the interval from (22 - 15) to (22 + 15), or
from 7 to 37. Assume that half of the frequency in the first and last class
intervals belong in this range from 7 to 37. Then the interval (mean - 2
standard deviations to mean + 2 standard deviations) contains 2+!, or 95%, of
the data.
FIGURE 11
These properties for the bar graph and the approximating smooth curve
demonstrate the following well-known properties of a normal curve (see
Figure 1 1).
Properties of a Normal 1. The area under a normal curve is 1.
Curve 2. The normal curve is symmetric about a vertical line through the mean of the set of
data.
3. The interval extending from 2 standard deviations to the left of the mean to 2 standard
deviations to the right of the mean contains approximately 95% of the area; the
corresponding interval extending 1 standard deviation on each side of the mean

contains approximately 68% of the area; the corresponding interval extending 3 stan-
dard deviations on each side of the mean contains 99% of the area.
4. If x is a data value from a set of data that is normally distributed, then the probability
that x is greater than a and less than b is the area under the normal curve between a
and b.
The curve in Figure 12 is the standard normal distribution. We use z to

represent the standard normal variable and v to represent the frequency.The
maximum value of the curve The standard normal curve
is attained at z = 0.
has perfect symmetry. Because of this characteristic, the mean and median
of the distribution have the same value
namely, 0. The range is not defined,
because there are values occurring as far out as you wish to go that is, the
curve never intersects the z axis.
FIGURE 12
The area underthe standard normal curve is Thus, to find the proba- 1 .
between z\ and zi we obtain the area under the curve between

bility that z is >
Z\ and Z2 (the shaded region in Figure 12). Table in Appendix B gives the 1
area under the normal curve less than or equal to z = Zi and greater than or ,
equal to z = 0. That is, the area indicated by the horizontal shading in Figure
12 is given in Table 1 at z Z\ The area from z = to z = Z\ is the same as
the probability that z is less than or equal to z\ and greater than or equal to 0,
or P(0 < z ^ Z\). The table stops at z = 3.09, because the area under the
curve beyond z = 3.09 is negligible.
The fact that the standard normal curve is symmetric about z =
means that the area under the curve on either side of is 0.5. This symmetry
allows us to compute the probabilities that do not specifically occur in the
table.
EXAMPLE 24 Find P(z < 1.84).
SOLUTION Since the area under either half of the curve in Figure 13 is 0.5000.
P(z < 1.84) = .5000 + P(0 < z s 1.84)

= .5000 + .4671
= .9671
FIGURE 13
For the normal curve, P(z = a) = P(z < a), and P(z s a) = P (z > a),
for In general, for distributions (called continuous distributions) like the

all a.
standard normal curve, the probability that z is less than or equal to a

number is the same as the probability that z is less than the number.
EXAMPLE 25 Find P(z s 1.2).
SOLUTION From Table 1 in Appendix B, we read that the value for z = 1.2 is
.3849. Thus,
P(z < 1.2) = .3849 + .5000 = .8849
Recall that the normal curve is symmetrical about z = 0. This fact is
important as we discuss areas under the curve. The fact that the standard
normal curve is symmetric about the origin means that the area under the
curve on either side of is the same. For example,
P(-1.05 < z s 1.05) = 2/>(0 < .x < 1.05)

= 2(.3531)
= .7062
Since the total area under the curve is 1, the area to the right of z = 1.66
is 1 minus the area to the left of 1.66. That is.
Piz s 1.66) = 1 -P(z< 1.66)

= 1 - [.5000 + />(0 < z 1.66)]
= 1 - [.5000 + .4515]
= 1 - .9515
= .0485
Sometimes we need to compute the probability that z is in a certain

range say, between 0.4 and 1.4 (see Figure 14). This probability is indi-
cated by P(0.4 < z < 1.4). It can be obtained by finding the probability that
z is less than 1.4 and subtracting from this the probability that z is less
than 0.4.
P(A< z< 1.4) = P(z< 1.4) - P(z < .4)

= .9192 - .6554
= .2638
FIGURE 14
EXAMPLE 26 Find the probability that the normal variable x, with mean 175 and standard
deviation 20, is less than or equal to 215.
SOLUTION 215-175 =
Z= 20
2
P(x < 215) = P(z = 2) = .5000 + .4772 = .9772
EXAMPLE 27 The grades on a certain test are known to be normally distributed, with mean
74 and standard deviation 8. What is the probability that a student will make
less than 58 on this test?
SOLUTION
z
= 58-74
o = -2
P(x < 58) = P(z < -2)

= P(z > 2)
= 1 - P(z < 2)
= 1 - .9772 .
= .0228
Thus, 2.28% of the students will make less than 58 on the test. Equivalently,
the probability that a student chosen at random will make less than 58 is
.0228>
EXAMPLE 28 The Iron Fist Security Agency has uniforms to fit men ranging in height from
68 to 74 inches. The heights of adult males are normally distributed with a
mean of 70 inches and a standard deviation of 2.5 inches. What percentage of
male applicants to Iron Fist can be fitted in their existing uniforms?
SOLUTION The z values that correspond to 68 and 74 are
-70 74-70
z = =
68
r-; -0.8 and z = r~l 1.6
P(68 < .v < 74) = P(-0.S < z < 1.6)

= P(-0.8 <z<0) + P(0<z<1.6)
= P(0 < z < 0.8) + .4452
= .2881 + .4452
= .7333
Hence, the probability that a given applicant can be fitted in a uniform is .73,
or 73% of the applicants can be fitted.
If x lies between (x x - crx and fix + cr ( , then, since
*--'-* = -1 & + * ~ *
and
we have
P(-l < z = 1) = 2P(0s z < 1)

= 2(.3413)
= .6826
SinceP(-2 < z < 2) = .9544 and P(-3 < z ^ 3) = .9974, the following
summary may be made for any normal distribution:
Norma/ Percents and If measurements are normally distributed with mean n,

Standard Deviations
Practice Problems > 1. On draw the histogram in Figure 9. Use X Min: 4.5, Max:
a calculator,
39.5, Scl: 5,Y Min: 0, Max: 6, Scl: 1. Now change the range to: X Min:
5, Max: 4 0, Scl: 5; Y Min: 0, Max: .07, Scl: .01. Use [Graph SHIFT 1
1
|
|
LINE EXE to draw the corresponding normal curve.
|
I | |
2. Write a program to graph
using A' Min: -3, Max: 3, Scl: I; Y Min: 0, Max: .5, Scl: .1. Compare with
the graph in Figure 12.
3. At the very exclusive Rochester Academy, the faculty wishes to admit

only the top 30% of those applying for entrance. The scores on the en-
trance exam are normally distributed with mean of 60 and standard devia-
tion of 10. What cutoff score does Rochester wish to use for admis-
sions?
Exercise Se! 7.4 Find the area under the standard

normal curve that lies between
the following pairs of values
ofz.
(a) z = to z = 2.40
2.4
(b) z = to z = 0.41
(c) z = to - = 1.67
(d) z = -0.36 to z = 0.36
Assuming these sketches repre-

sent the standard normal curve,
compute the shaded areas.
-1.68 1
3. Find the following probabilities

from Table 1 in Appendix B.
(a) P(z 2 -2.1)
(b) P(z> -1.4)
(c) P(z s 0.1)
(d) P(z< -1.6)
(e) P(z> 1.5)
(f) P(z > 2.4)
(g) P(z> -2.1)
(h) P(z > -1.8)
(i) Pil.3^ z < 2.4)
(j) P(2.1< z< 2.8)
(k) P(-l.2 < z = 0.3)
(I) P(-2.6< z < 1.4)
4. If .v is a variable having a normal

distribution with v = 12 and sx =
4, find the probability that x as-
sumes the following values.
(a) ,v
<
Applications
(Business and 10. Sales. It is known from experi- their tires is 55,000 miles (with a
Economics) ence that the number of tele- standard deviation of 4000
phone orders made daily to a miles), and they also know that
company approximates a normal occasionally a tire fails in less
curve with mean 350 and stan- than 50,000 miles. What per-
dard deviation 20. What percentage of the tires will fail be-
centage of the time will there be fore 50,000 miles?
more than 400 telephone orders
per day?
12. Quality Control. The life of a
certain brand of batteries has a
11. Quality Control. The Goodbond mean of 1200 days and a stan-
Tire Company manufactures a dard deviation of 100 days. If the
superior quality tire (sold at a manufacturer does not want to
superior price) that they guaran- replace more than 12% of the
tee for 50,000 miles. They know batteries, for how long should
that the average life of one of the batteries be guaranteed?
and Life Sciences) 13. Radar. Radar is used to check mally distributed with mean 177
the speed of traffic on Interstate centimeters and standard devia-
75 north of Atlanta. If the mean tion 4 centimeters.
speed of the traffic is 70 miles
(a) What percentage of the men
per hour with a standard devia-
are between 173 and 181
tion of 4 miles per hour, what
centimeters in height?
percentage of the cars are ex-
(b) What percentage of the men
ceeding the legal speed of 65
are between 169 and 185
miles per hour?
centimeters in height?
14. Diets. Young rabbits placed on
16. Tests. The scores on the en-
a certain high-protein diet for a
trance exam for the Kentucky
month show a weight gain with a
Police Academy are normally
mean of 120 grams and a stan-
distributed with a mean of 60
dard deviation of 12 grams.
and a standard deviation of 6.
(a) What is the probability that a What is the probability that a
given rabbit will gain at least randomly selected test score will
100 grams in weight? lie between 60 and 75?
(b) If 15,000 rabbits are placed

17. Grades. The grades in a certain
on this diet, how many can
class are normally distributed
be expected to gain at least
140 grams?
15. Heights. The heights of men in

a certain army regiment are nor-
with mean 76 and standard devi- 19. Aptitude Tests. The average
ation 6. The lowest D is 61, the time required for completing an
lowest C is 70, the lowest B is aptitude test is 80 minutes with a
82, and the lowest A is 91 . What standard deviation of 10 min-
percentage of the class will utes. Assuming that the lengths
make A's? What percentage will of times necessary for complet-
make B's? C's? D's? F's? ing the test are normally distrib-
uted, when should you stop the
18. Achievement Tests. A nationally test to make certain that 90% of
administered achievement test is the people taking it have had
known to have a mean score of time to complete the test?
500 and a standard deviation of
100. What is the probability that
a score is less than 300?
Answers to Practice t> 1. Program: |

SHIFT |
|
MODE |
[x~\ |
MODE j Q] 7 SHIFT] [ScT| |
EXE |
|
Problems |XE 1
5 [DTI IQ JDT |[DT| 15 1
SHIFT 1
[T|4 [lyfl 20 [SHIFT |T| 6| DT 25 |
1
|
SHIFT] Q] 4 |
DT |
30 |
DT DT 1
1
|
35 |
DT 1
1 Graph |EXE| \
(Put in new range.) [Graph 1

1 SHIFT 1
1 Line EXE Compare with
|
1 1
|
.
Figure 10.
2. 1
Graph |
0J_E] I V I 2 HI H BED LH ALPHA l l
H CEI
r^rn iEXEi
3. The cutoff score is 65.3.
7.5 Estimating Using Sample Measurements
Overview Professor Larkin in the education department of Lagoon College wishes to determine the
average ACT score of the students. Not possessing the time or energy to collect a score from
every student, he randomly selects a sample group of 30 students and determines their scores.
The average of these 30 scores is 21. Professor Meuler in the psychology department also
wishes to obtain this information. She too interviews a sample of 30 randomly selected students
and determines that their average ACT score is 19.7. Which professor has the correct informa-
tion? In this section, we will study
The central limit theorem

Confidence interval for the mean
The most important techniques of statistics are those that allow us to

determine information about a large collection of data by examining a small
collection of data chosen from thehave learned that the

large collection. We
large collection of data is from
called the population; a small collection taken
a population is a sample. In the discussion above, the two professors were
attempting to determine the mean (or average) of the ACT scores of the
whole student body by looking at a sample of only 30 such scores. In both
cases the professors got a number that was an approximation of the mean
itself. The best way to give information about a population mean obtained
from a sample mean is to give an interval in which the population mean

probably lies. We now describe the tool that allows us to determine this
interval.
Suppose that from a population we have collected all possible samples
of sizen. Suppose further that we have computed the means x of all these
samples. We have then created a new population (consisting of sample
means) that is called the sampling distribution of the mean. It can be shown
that this sampling distribution isapproximately normally distributed; the
mean of this sampling the mean of the population (designated
distribution is
by /j.); and the standard deviation of the sampling distribution is oVVn,

where a is the standard deviation of the population. Thus, if J is a variable
representing the sampling distribution, then
r lVr~i
is the standard normal variable.
Centra/ Limit Theorem If x (the mean of a sample of size n) is considered as a variable in repeated random
sampling from a given population, then the following statements are true.
1. The mean p* of the variable x is equal to the mean /x of the population
2. The standard deviation o> of the variable x is equal to the standard deviation <r x of the
population divided by the square root of the number of items in each sample
Vn
3. The variable z, where z is given by
o-jVn
is approximately a standard normal variable if n is sufficiently large.

Estimating Using Sample Measurements 453
One immediate difficulty is that if we are trying to estimate the popula-

tionmean, we probably do not know the population standard deviation.
Then our best estimate of the population standard deviation will be the
standard deviation of the sample. In fact, in many confidence-interval prob-
lems we are forced to use the standard deviation of a sample.
Now let us investigate the probability that z lies in a given interval for
the standard normal curve. For example, what is the probability that z lies
between -1.96 and 1.96?
P(-l.96 < z s 1.96) = P(z = 1.96) - P(z s -1.96)

= />(z< 1.96) - P(z s 1.96) Why?
= .9750 - [1 - P(z < 1.96)] Why?
= .9750 - [1 - .9750]
= .9750 - .0250
= .95
As noted in Figure 15(a), 95% of the time z lies between -1.96 and 1.96.
EXAMPLE 29 Show that z lies between -2.58 and 2.58 approximately 99% of the time.
(See Figure 15(b).)
-1.96 1.96 -2.58
(a) (b)
FIGURE 75
P(-2.5S < z < 2.58) = 2P(0 < z < 2.58)

= 2(.4951)
= .9902, or approximately 99%
Now, starting with a mean of a random sample, we wish to find an
interval in which we can be reasonably confident that the actual population
mean lies. More precisely, we wish to find an interval so that we are 95%
confident that the actual population mean is in that interval. Such an interval
will be called a 95% confidence interval, and the endpoints of the interval are
called 95% confidence limits. Consider a sample mean x from a sample of
size n. Suppose the population mean is /i, and the estimate of standard
deviation is s. Since z = (jc - yi)l(sl\fn), then 95% of the time (x - fi)/(s/
\rn) is between -1.96 and 1.96.
x - fl
1.96 < 1.96
slVn
What does this statement say about the population mean?
X x - ii
fJL
1.96 > -1.96
s/Vn ;/Vn
x- fi< 1.96 1- x - p. > - 1 .96
f+l% 1.96
fe
(jl > .v - 1 .96 M <.v+ 1 .96 ( -4=
fe)
Hence, we may be 95% certain that the population mean is between
r
x - 1.96(s/V/2) and x + \.96(sl\ ii). Similarly, we may be 99% confident
that the population mean is between x - 2.58U/V/7) and x + 2.58(^/V).
(Can you verify this?)^
Formulas for specified confidence intervals may be summarized as

follows.
Confidence Intervals lor 99% confidence:

the Mean of a Population
Estimating Using Sample Measurements 455
SOLUTION The 95% confidence interval for fi, is from 870 1.96(20)/V64to
870 + 1.96(20)/V64, or 865. < Ml < 874. 9> 1
EXAMPLE 32 In order to analyze plant efficiency, an estimate of the average daily output
at a certain factory is needed. A random sample of 36 days yields an average
output of 600 items per day, with the estimate of the standard deviation being
10. What is the 90% confidence interval for daily production?
SOLUTION
1 .65 I

iV///
to v + 1 .65
\Vn
,
600
1.65(10)
V36
1=^ to
, nn
600 +
1.65(10)
L
V36
^
597.25 to 602.75
We can be 909? certain that the average daily production for the year is
between 597.25 and 602. 75>
Exercise Set 7.5 Find 807c, 907c 957c, and 997c confi- 6. We have determined that - =
dence intervals for fix from the sta- -1.96 to z = 1.96 is the 95%
tistics given in the following sam- confidence limit. Find z for the
ples: following confidence limits:
1. 800, s, = 60. n = 100

(a) 60%
(b) 75%
100. 10. n = 36 (c) 85%
(d) 98%
3. v = 40. i, = 2, n = 400
7. In the process of estimating a
4. x = 10. sx = 1, /; = 64 population mean from a sample
mean, it is desired that the confi-
5. Each person in a random sample
dence interval be .6 in length. If
1
of 81 people vacationing in Wyo-

the population standard devia-
ming was asked how much
tion is 4, how large a sample
money was spent during the trip.
The average amount for the
must be taken to obtain a 95%
sample was $361 with s = $48.
confidence interval? A 90% in-
x
terval?
What is a 90% confidence inter-
val for the average amount spent 8. It is desired to use a sample
by all tourists in Wyoming? mean .v to estimate the mean of a
normally distributed population mean of a

probability that the
with an error of less than 0.6 (the random sample of 100 measure-
length of the confidence interval ments will exceed 1640?
is 1.2). If the standard deviation
11. In Exercise 10 find the probabil-
of the population is 4, how large
ity that the mean will exceed
a sample must be taken for a
1620.
95% confidence interval?
12. A given population has a mean
9. In Exercise 8 how large a sample
of 200 and a standard deviation
would be necessary for a 99%
of 20. What is the probability
confidence interval?
that the mean of a random sam-
10. Suppose a population mean is ple of size 100 is in the range 198
1600 with cr, = 100. What is the to 200?
Applications
(Business and 13. Manufacturing. The average into paper bags. The company
Economics) daily production of automobiles wishes to estimate the true mean
at the Bowling Green plant over weight of the bags. A sample of
the last 100 days is 240 cars with 36 bags yields a sample mean of
a standard deviation of 20 cars. 24.75 pounds and a sample esti-
Find a 95% confidence interval mate of the standard deviation
for average daily production i x
. of 0.25 pound. Find a 99% confi-
dence interval for the true mean
14. Manufacturing. A sugar manu-
weight of all the bags of sugar.
facturing company packs sugar
and Lite Sciences) 15. Diet. A doctor has been study- confidence of 95% the mean
ing the effect of a new diet on 36 weight of all babies born in this
infants. In a 1-month period, the hospital in 1980.
mean gain in weight is 3 10 grams
17. Commuting Students. In a sam-
with a standard deviation of 100
ple of 49 commuting students at
grams. Find a 95% confidence
Gordy College, the average one-
interval for the average gain in
way distance traveled to school
weight of infants on this diet.
each day was found to be 10.2
16. Weight. A random sample of miles. The sample estimate of
size 36 is taken of the weights of the standard deviation was 1.5
babies born Happiness Hospi-
at miles. Give a 90% confidence in-
tal during The sample
1980. terval estimating the average
mean is 7.6 pounds, and the one-way distance traveled by
sample estimate of the standard commuting students at Gordy
deviation is 0.8. Estimate with a College.
A Comparison of the Hi iiomi.il and Normal Distributions
A binomial distribution is a discrete distribution with two possible outcomes,

which we call success and failure. The binomial distribution function is a
discrete distribution function because it is defined for only integral values, 0,
1, 2, 3 /;. For each variable, x, is given a probability
C(n, x) = p*q"~ where q = p - I p is the probability

of success
The mean and standard deviation for this distribution are given by
M. = "P
<j x
= V npq
The normal distribution function is a continuous distribution function
(the variable can take on all real values) defined by
f(x) = L=*-u-M,*2ri
o-, V2tt
where p., is the mean of the distribution and cr, is the standard deviation. By
making the transformation z = (x p )/cr, we obtain the standard normal
y
,
distribution function with mean and standard deviation I
To demonstrate the use of these two distribution functions, consider

the following example. Seventy percent of the participants at a state Repub-
lican convention are conservatives. If 12 people are chosen at random to be
the rules committee, what is the probability of getting 8, 9, or 10 conserva-
tives?
P(8) = C(12, 8)
(.7) (.3)
8 4
p = .1
9 3 q = .3
P(9) co; 9)
(.7) (.3)
l0 2
P(\0) en: 10) (.7) (.3)
So the probability of getting 8, 9, or 10 conservatives is .6386.

For many practical problems the methods of using the binomial distri-
bution become very cumbersome. For this reason practical problems are
usually worked with a normal curve approximation to the binomial distribu-
tion. The mean and standard deviation of the preceding distribution are
p = up = 12(.7) = 8.4
x
o-, = Vnpq = V'12(.7)(.3) 1.587
The following diagram shows the normal curve approximation to the histo-
gram representing the binomial distribution.
The probability of 8, 9, or 10 conservatives would be the area under the

normal curve from 7.5 to 10.5. The z scores for these values are
7.5 - 8. 4 8.4 is the mean and 1.587 the

= -.57
1.587 standard deviation of the
normal distribution.
and
10.5 - 8.4
= 1.32
1.587
The area under the normal curve from z = .57 to z = 1.32 is .6223. This
normal distribution area is a good approximation to the answer, .6386.
Consider another example. During inspection of 1000 welded joints
produced by a certain machine, 100 defective joints were found. Consider
the random variable to be the number of defective joints that result when 50
joints are welded. Compute the mean value of .v, the standard deviation, and
the probability of getting more than /i,, + 2o\ defective parts, where fix + 2ax
stands for 2 standard deviations above the expected value.
100 _ J_
1000
" 10
Thus,
Mv = np = 50 ( ) = 5
a,. = V npq (*>(> 2.12
px + 2o- v
= 5 + 2(2.12) = 9.24
Therefore,
P(x > 9.24) = P(IQ) + P(\\) + P(\2) + + P(50)
This requires a great deal of computation using binomial probabilities. How-

ever, using the normal distribution
9.5 - 5
z = = 2. 2
2.12
and PL: > 2.12) = .017.

Thus the probability of 10 or more defective parts is approximately
.017.
The normal curve approximation to the binomial distribution is usually
quite accurate, especially when both up and nq are 5 or more.
As you read the following review list, be sure you understand the meaning
(or definition) of each of the terms. If you are not completely satisfied with
your knowledge of the term, review the pertinent material.
Important Frequency (408) Circle graph (415)
Terms Frequency distribution (408) Mean (423)

Grouped frequency Median (424)
distribution (408) Mode (425)
Class marks (413) Variance (432)

Bar graph (411) Standard deviation (432)
Histogram (413) Normal distribution (442)
Line graph (412) Sample estimate (452)
Confidence interval (453) Central limit theorem (452)
Frequency polygon (414) Z score (437)
Chebyshev's theorem (437)
If you do not immediately recognize a formula in the following list of formu-

las from this chapter, look it up and review its use.
Important .V, + -V 2 + .v, + + x

Formulas
, (.V, - ,Y)
:
+ (*2 - .V)
:
+ + (.V - Xf
JCl/l + *2/2 + ' + Xm f,

A + fl + '
+ fm
sx = Vvariance
X- Z 7= < jU* <X + Zi
Sjc 2 - r* 2
n -
A
Chapter Test
Given the values 3, 7. II, 15, 19, 23,
1. Find the mean.
2. Find the median.
3. Find the variance.
Problems 47: For the frequency distribution
Interval
Appendix A
Inexpensive calculators with the potential to sketch quickly the graph of a function
are changing the way professors teach and students study the material in this book.
Procedures are given below for using one such calculator, the Casio fx-7000G. I do
not advocate the use of any particular calculator; however, the calculator discussed
here is the least expensive graphing calculator currently available. I expect this
situation tochange within the next few months. To illustrate the use of this calcula-
tor, we use the keyboardin the picture to work examples involving the sketching of
curves in the first part of the book. From these examples, you should be able to use
the graphing calculator throughout the remainder of the book.
Two types of graphs can be generated using this calculator: built-in function
graphs and user-generated graphs. The calculator contains the following built-in
graphs for this course.
sin cos tan x2 log In
10
We can draw such graphs as follows:
|
Graph EXE sketches v =
|
Graph EXE sketches y = e*
Any time a built-in graph is executed, the ranges are set to their optimum
values. User-generated graphs are not set automatically: instead the user must de-
cide upon the range. The parts of a graph outside of the selected range do not appear
on the display. The range contents are given in the fi gure: the maximum and mini-
mum on each axis and the scale on each axis. Press the |
Range key and the following
|
appears.
463
464 Appendix A
Appendix A 465
Range
466 Appendix A
Chapter 1
v = x + 4
Figure 10
|
Graph |
[FT] |
ALPHA |
[FJ [T] 4 f~EXE~1
H 4-
Figure 11
|
Graph |
|
ALPHA |
\Y\ [7] 1 |
EXE |
V = X + 1
Two graphs can be graphed on the screen at the same time and the calculator
can be used to obtain an approximate solution.
Figure 14
320 - Ax
p = 20.v
7--
/
/
/
/
/
/
-f 1
Chapter 1 467
First change the range on x to 2 6 with a scale of 1 , and on y to 80 y 120
with a scale of 10.
|
Graph | Q] 320 Q 4 |
ALPHA |
5
I
SHIFT [J] 1
Graph |
20 |
ALPHA |
{Y\ |
EXE | |
EXE
To read the point of intersection, use |
SHIFT |
|
Trace |
. Locate the blinking pixel
at the leftof the screen. Use the arrow ke ys and to move it to the point of g] g]
intersection. Then enter SHIFT X<-* Y and read the y coordinate of the point of
|
1
1
|
intersection. For greater accuracy we will use an automatic zoom-in feature. Press
|
SHIFT |T| (the key). You can also zoom in by changing the range settings, but
| [f\
the preceding procedure is much faster The fi rst graph ap pears automatical ly. The .
second graph appears when you press EXE Again use [SHIFT] Trace Now. |
|
. |
|
.
move the pixel to what seems to be the intersection and read x and y. This whole
process can be repeated until you obtain the accuracy you desire. In two repetitions,
it was found that x = 5 and p = 100.
Figure 22
In this figure are graphs y = 4.v, y = 2.v, y = v, and y (l/2).v in this order.
|
Graph 4 1
|
ALPHA |
\x\ |
SHIFT |
[d]
|
Graph |
2 ]
ALPHA \Y\ |
SHIFT |
[d]
|
Graph | |
ALPHA |
[x] [SHIFT] |T1
[Graph |
.5 |
ALPHA |
|T] |
EXE | |
EXE | |
EXE |
Did you get the lines in the figure?

468 Appendix A
Graphing Statistical Data
The graphing calculator is very useful when working with statistical data. For exam-
ple, suppose you have the following set of data with frequency.
Class Interval
Graphing Statistical Data 469
10 [
DT [
|
DT | |
DT |
|
Note 3 DT's]
10 |
SHIFT |
[7] 3 |
DT |
Both of these procedures will be used in the following program:
[DT] 10 |
DT |
|
DT |
|
DT |
20 |
DT | |
DT |
30 ]
DT |
|
DT |
40
\W\ [DT] [DT] 50 |

SHIFT] Q] s
- [DT] 60 I
SHIFTJ HI 6
|W] 70 I
SHIFT I Q] 8 Q5t] 80 |
SHIFT) \T\ 15 [W] 90
|
SHIFT | Q] 9 [DT] 100 [DT] [DT] |
Graph | |
EXE |
Do you get the following graph?
To overwrite a line graph on the histogram, use
|
Graph | |
SHIFT |
|
Line EXE
Statistical computations also use the statistical mode
1
MODE |
|"x~| |
SHIFT |
[ScT] |
EXE |
The last three instructions clear the statistical input data, use the class memory. To
mark of an interval and the following procedures for frequency. Suppose a class
mark is 10 and the frequency of 10 is 3. We can enter this statistical information as
either
10 |
DT |
|
DT |
|
DT |
|
Frequency ofT|
10 |
SHIFT | Q] 3 |DT|
To obtain the sum of all frequencies entered, use
| ALPHA |
[W] [EXE
470 Appendix A
and to obtain the arithmetic mean of the data, use
|
SHIFT |
[JJ |
EXE |
To obtain the standard deviation s, of a sample use
SHIFT .v<7-, EXE

We illustrate these concepts by finding /;, the arithmetic mean x, and the
standard deviation .v, of the following sample:
Appendix B
Tables
TABLE 1
472 Appendix B Tables
TABLE 1
AREAS UNDER THE STANDARD NORMAL CURVE A(z ) (cont.)
zo
Appendix B Tables 473
TABLE 3
COMPOUND INTEREST
i = 1% (interest rate per period) n = number ot periods
n
TABLE 3
COMPOUND INTEREST (cont.j
i = 2% (interest rate per period) n = number of periods
n
TABLE 3
n
TABLE 3
n
TABLE 3
n
TABLE 3
COMPOUND INTEREST (cont.)
i = 8% (interest rate per period) n = number ot periods
n
TABLE 4
ANNUITIES
(1 + / )" - 1 /)"
s ^\i = :
flJfli
n
TABLE 4
ANNUITIES (cont.)
I= 2% (interest rate per period) n = number of periods
(1 - ty - l 1 - (1 + I)-
am =
n
TABLE 4
ANNUITIES (cont.)
i= 3% (interest rate per period) n = number of periods
SQi =
+ (1

i)
;
n
~ 1
fl S|/
1
- (1
n
TABLE 4
ANNUITIES (cont.)
I = 4% (interest rate per period) n = number ot periods
(1 + /)" - 1 1 - (1 + /)-
sm = ;
am =
n
TABLE 4
ANNUITIES (cont.)
= (1 + /)" - 1 1 - (1 + )-
s n\i am
n
TABLE 4
ANNUITIES {cont)
(1 + /)" - - (1 + /)"
ssi.-
= ;
1
ami
1
n
TABLE 5
VALUES OF EXPONENTIALS
X
TABLE 5
VALUES OF EXPONENTIALS (cont.j
X
Answers to Selected
Exercises and
Chapter Tests
Chapter 1
488 Answers to Selected Exercises and Chapter Tests
19. Undefined 35. H' = {A, F, G}
23.
27. 2.vy
29. -1
31. (a) 17.2129 (c) 355.4321
32. (a) .0756 (c) 4.89552 37. $66.92
33. (a) 7000 (c) .2 39. .5, .36, .015.
Exercise Set 1.3

(page 26) 1. x = 5 13. (3, )
Check: 2(5) -7 = 3 15. (-, -21)
10 - 7 = 3
3 = 3 -)
3. x = 3 A P
Check: 4.v -7 = 5 19 Pt 4
Pt
4(3) -7 = 5
12 - 7 = 5 21. y mx
5 = 5 S -
23. r 5 4
5. x = -
Check: 25. -v = 2
4 - 2(-l) = [8 + 3(-D] + 1 Check:
4 + 2 = (8 - 3) + 1 -[2(2) - (3 - 2)]= 4(2) - 11
6 = 5+1 -[4 - = 8 - 11
1]
6 = 6 -3 = -3
7. x = 1
+ = 6+1 27. x =
Check: 2(1) 5 3
2 + 5 = 7
Check:
7 = 7
9. x = 5
' -"-)
Check: I - 3 = -2 4
= 8
9 9
- 3 = -2
-2 = -2
9 9
11. v = -4
-4 29. a- = 9
1
Check: -=- _ "
2 9-5 9-9
3 5 Check: I +
4 12
ill
_5_
15
15
-20
-17
15
15
+
3
T5
4
4
I
a
1 +
1
Chapter 1 489
31. a = 5 or x = 3 39. (-5. 9)

Check: |4 - 5| = 1, |4 - 3|
= 1
|-1| = 1 |1|
= 1
1 = 1 1
= 1
41. (-x, -3]

33. .v = -2, a = 1
Check: 1
|2(-2) + 1| = 3. |2(1) -+-1| = 3

|-4+ 1| = 3 |2 + 1| = 3
|-3| = 3 = 3 43. P =
|3|
1 + rt
3 = 3 3 = 3
1 + 3jc
35. [-6. 1] 45. y
-2z 3
47. x < 1.696
49. a 6.6402
51. .v = .2003
37.
-#-12
Exercise Set 1.4

(page 51) 1. 10 13.
3. Width = 4.4 inches

Length = 14.6 inches
5. j gallon
7. (a) 9375 ties
9. $22,727.27 and $37,727.27
11. Brand A = $1.55

Brand B = $1.30
Brand C = $2.23
Exercise Set 1.5

(page 41) 1. (a)
(2,4)
-1
-2 +
2. (a) \.yj
3. (a) (0.2), (l,|), (-1,3
< c ><o,5,,(.,f),(2,f)
-1,
(e) (0, 1), (l.^j (2,D,(4,4).

(-4, -2)
5. (a)
y= 2a- - 3 (g) y
3"
2--
1--
1
1 2 3 -
y + 1 =
-2
-3 +
(c)
4-
kk 3-
(0,2)
-
I
(3,0)
1
I
2
XH
3
I
'
,
-1-
-2 +
Chapter 1 491
6. (a) a intercept: -3
v intercept: 2
(c) v intercept: 6 11. w < 3.t

y intercept:
-
7. v = 2.v -3, -1 s.v < 4

15. y.
5-
Chapter 1 493
y
11. (a) y = 4x - 5
3 +
x
1 2 3
-l
2y= 1 -x
-2 +
17. Slope = 3
y intercept: 1
19. Slope = =
y intercept: =
y.
3.v-2v = 5
3--
2--
13. Slope = -2 1
--
y intercept: 1
-2
(O.-f)
-4
20. (a) F (c) F
1 5
21. v = rX +
2
23. v="3*"37 1
= = 25. y = 4.v - 3
15. Slope y intercept: =
27. .v = 1
28.
Chapter 1 495
7. 3v + 2.v - 19 =
9. (a) y < 3x + 2 (c) 3* + 2y = 1 (e) x + y < 5
H 1 (-
-1/
I
-2"
(a)
MM
17 . Pll + Ell = , Divide both sides by M
19. t
=
20. (a) 29.18209003
21. Wife receives $12 000 ?
Each child receives $4000

23. 450 27. $1.04; $2.80
25. 2 hours 29. 69
Chapter Test
(page 59) 1. 2x + 3v = 6
Chapter 2 497
Chapter 2
Exercise Set 2.1

(page 68) 1. 2 x 3, a 12 = 1 -12
25. X =
3. 3 x 3, c 32 = 2 -1
5. Impossible 26. (a) [150 200 300 100]
7. [18 24 30] 27. (a) (A) (H) (C)
[ 6 12 18J W34 MOO 50 200"
9. Impossible, the two matrices are not G47 60 40 70

the same size.
B71 13 40 24
11. 20 10 15 15
-5 10 -5
13. 18 -7 17
1 18 19
-9 14 -11
IS. -2 -1
17. 35 -5
-2 28
"2
19. [2
3
4
_5
21. x = -4, y = 8, z = 20
23. x = 0, y = 0, z =
Exercise Set 2.2

(page 78)
37. 1
1
1 2
39. [6
41.
Chapter 2 499
2. (a) (2, 3)
(c) In addition and subtraction you
get = 0. The lines coincide and
there are an infinite number of so-
lutions, (one possibility)
(e) (1,3)
(g) In addition and subtraction you
get = 1. This is impossible so
there are no solutions,
(one possibility)
(i) (1, 2)
3. (a) x = 5, y = -1
(c) x = 12, y = -5
(e) x = 0, y = 2
,
4 (a) x
4. = _ jo
- , v
y -
= n*
(g) 133" 133
(c) x = 50, y - 3
5. 93, 82
7. (a) x = 1.3417, y = .0761

(c) x = il,y = 103
9. 32 pounds of $3.00 per pound candy
48 pounds of $4.00 per pound candy
11. 1400 items at a cost of $350
13. Yes. The break-even point will be

reached at a lower level of produc-
tion.
15. Number of women: 360

Number of men: 640
Lines are parallel: no solution

Exercise Set 2.4

Chapter 2 501
1 2
1 3
5 7
3 4
2. (a)
(c; 4 -19]
L-i 5j
3. 2x - y = 4
3a- + 4y = 1
5. x + 3y + 2z = 5
-Jt + 2y + z = 1
y + 2z = 4
3l
11. (a) A "!]

[-1
x = 4, y =
(c) A
x = 3, y = 2
. (a)
Exercise Set 2.7

(page 132)
T20
= (rounded)
|
28 I
7. X -]
Chapter 3 503
6. (a) A B C D (c) Infinite number of solutions

A TO 1
0" (e) No solution
B 10 10 11. (a) Unique solution
C x = 1 y = 1 ,
D |_1 I
(c) Infinite number of solutions
7. (a) No solution 27 - 8c 16 + 3c
(c) Infinite
If z =
number of solutions
t,y = s, then
*
:
J," .y = i- .?
12. (a) Infinite number of solutions

x = 25 + 3r - 3
x = 1, y = c 2,
2c+
8. (a) x= \,y = \,z= 1
z = -c + 3, t = c
9. (a) x= l,y = \,z= 1
10. (a) No solution
Chapter Test
(page 139) (b)
8 6
-4 -2
1 2
1
3J
7. * = 9, y = 7
8. a: = 9, y = 7
9. x= l,y = l,z= 1
10. $1000 at 6%, $3000 at 8%, $6000 at

10%
II.
2. x = 2,y = \
3. a = 5, b = 9, c = 3, d = 2. 14,
/=20
Chapter 3
Exercise Set 3.1

(page 148)
3--
v = 3
2
-\ I
H
12
I
3
I I
4 5
Unbounded; corner point: (0, 0) Unbounded; corner point: (0, 3)

Unbounded; comer points: (0, 0),
(3,3)
Unbounded; corner point: (0, 0)
7. >'i
Unbounded; corner point: (0, 0)
13.
Bounded; corner points: (0, 0),
(0, 3), (3, 3)
Bounded; corner points: (1, 0), (2, 1)
(0, 1)
Chapter 3 505
Bounded; corner points: (-s-, =rj, 19.
(14, 10), (2,0), (4,0)
17. >*
.v +y=2
(0, 1), (1,0)

(4,0), (3, -1)

Corner points: (0, 0), (4, 0), (0, 4)

Comer poims .
(Q Q) (5Q Q) ^ 5fJ)
23. * + >><$30.000;*>2y;*2 0;y2 27i;[ > > 7, , + , < 50; 7. +

, y
3y < 210
v.
Corner points: (0, 0), ($30,000, 0)

($20,000, $10,000)
Corner points: (0, 7), (0, 50),
(15, 35), (27, 7)

Chapter 3 507
Exercise Set 3.2

(page 157) l.
Corner points: (0, 0), (0, 7), (2. 5),
Corner points: (0, 0), (0, 5), (4, 0), (3. 3). (4. 0)
(3, 3) Maximum value of F is 245 at (0, 7).
Maximum value off is 180 at (3. 3).
Maximum value of F is 52 at (3, 2).

Minimum value of F is 12 at (0. 4).
7.
Minimum value of 20 at (0, 8). Maximum value of P is 245 at (0, 7).
No maximum value. Minimum value of P is 50 at (2, 0).
Maximum value of P is 180 at (3, 3).
Minimum value of P is 50 at (2, 0).

13. Chapter 3 509
Maximum value of P is 400 at (8, 0).

Minimum value of P is 40 at (0, 2).
Exercise Set 3.3

(page 164) 1. (a) Type I Type II 2. (a) Type I Type II
2 3 3
2-14 + 3
13= 67
14x + 13y = P
4 2
4jc + 2y < 12
3 3
3jc + 3y < 12
Corner points: (0, 0), (0, 4), (3, 0),
(2,2)
3. (a) Refinery 1
Chapter 3 511
Exercise Set 3.4

(page \77) 1. 3.v + 2y + r = 8
lv + 4y + i = 8
-14* - I2> + P =0
jc = r = 8
v = ^ = 8
/> =0
3. 2.v + 4r + = ,s 7
3.v + y + 2r = 5
4.v - 3;- + P = 12
.v = y = 5
r = 5 = 7
P = 12
5. .V
27. x y r s
10 10
J 1
110
Exercise Set 3.5

(page 186)
Exercise Set 3.6

(page 194)
>2 P
2
3
3.
Corner points: (3, 0), (5, 0), (0, 4),
(0, 8); P is a minimum of 12 at (3, 0).
/21 9\
Corner points: (0, 2), (0, 9), hr ,
g] ;
P = 16 at (0, 2) is the minimum.

Chapter 3 515
-8 - k k
-I I
+ k -7 - 2/ k -4A
11. -1 1
I -1 1
- k -6 - k A
(No solution exists.)
13. - 1 3 1
2 1
-6
15. P 80 at (0. 8)
17. C 12 at (3. 0)
19. P 16 at (0. 2)
21.
Review Exercises
(page 204)
Chapter Test
(page 210) 1
Corner points: ( 0, ~ , (
J
bounded
160
3
imum value at (0, 0)
2
Chapter 4 517
6. (a) $1255.56; $4255.56 19. $43.01

(c) $1288.51; $4288.51
21. 23 days
7. (a) A = $4880 (c) A = $124
23. $8074.76
8. (a) 9% (c) 16% 25. $486.11
9. $2380.57
27. (a) $71,633.91 (c) $6.27
11. $4391.89
28. (a) $86,357.00 (c) $7.56
29. (a) $103,749.70 (c) $9.08

"--A 30. (a) 24 years (c) 15 years
15. 9.5 years
31. 88,815
17. $133.33
Exercise Set 4.2

(page 230) 1. Geometric progression with r = 3 13. (a) $3105.85
3. Not a geometric progression (c) $3262.04

(e) $3319.46
5. Not a geometric progression
15. 10.52%, compounded annually
7. Not a geometric progression
17. $33,194.62
Exercise Set 4.3

(page 236) 1. $1318.08. Amount of an annuity of At the end of 20 years he will have
$100 a year at 6% compounded an- $91,523.93.
nually for 10 years, (one possibility)
17. His deposit is $101.24.
3. $4842.98. Amount of an annuity of
$200 a year at 8% compounded an- Year Deposit Interest Amount
nually for 14 years, (one possibility) $101.24
1 $0 $101.24
5. $1318.08 2 $101.24 3.04 205.52
7. $2687.04 3 $101.24 6.17 312.93
9. $122,445.22
19. $12,283.97: $2093.
11. (a) $45,761.96 (c) $7203.66
13. $5866.60
15. (a) Year Deposit Interest Amount
Exercise Set 4.4

(page 242)
s.
Chapter 5 519
17. $255,470,000,000 (rounded) 21. $4136.65
19. $12,576.99 23. $106,366.28
Chapter Test
(page 254)
12. (a) (c) 1 (e)

^
20. {(1,
(2,
1),
2),
(1,
(2, 3),
2), (1,
(2,
3),
4),
(1, 4),
(3, 1),
(2,
(3,
1),
2),
(3, 3), (3, 4), (4, 1), (4, 2), (4, 3).
(4, 4)}
K) 5
14. (a) No, probability cannot be nega- (a) =
16 8
tive.
(c) Yes 2_ = 1
(c) ~~
(e) No, probability cannot be larger 16 8
than 1.
(g) No, probability cannot be larger 21. (a)
than 1.
15. {HI, H2, H3, H4, H5, H6, Tl, T2, (c)
T3, T4, T5, T6}

100
16. (a) {R, B} 23. (a)
108
(c) {RRR, RRB, RBB, BBB}
17. (a) No, you can get heads. (C) =
T08
(c) No, 4 is not a possible outcome.
25, (a) {(Z, V, W), (Z, V, X), (Z, V, Y),
18. (a) {(1, 1), (1,2), (1,3), (1,4), (2, 1),
(Z, W,
X), (Z, W, Y), (Z, X, Y)}
(2, 2), (2, 3), (2, 4), (3, 1), (3, 2),
(c) {(W, X, Y), (W, X, Z), (W, Y, Z),
(3, 3), (3, 4), (4, 1), (4, 2), (4, 3),
(X, Y, Z)}
(4, 4)}
=
19. (a) j-
4
,c)
h \
(e)
T4"
Exercise Set 5.2

(page 272) 1. (a) 24 7. 2
3. 72 9. 24
5. 18
10. (a)
^
11. (a) (C)
T6 T6-
U. (a)
(C)
35
13. (a) 358,800

(c) 456,976
14. (a) -^
15. (d)
40 2
16. (a)
100
(O -2L
vv-'
inn
100
= li 1
(e)
il
100
Chapter 5 521
17.
Exercise Set 5.3

(page 281)
13. (a) 10,000 21. (e)
23. (c)
14. (a)
|| 25. 30
26. (a) 1140
(c)|
16. (a)
17,576
27. (a) 1900
63 (c) 648
17
125
28. (a) 120
19. (a) 1716
29. 120
(c) 725,010
Exercise Set 5.4

(page 294) 1. .7 9. (d)
3. (a) Getting fewer than 2 heads 11. 85%

(c) Getting at least 2 tails or no tails 12. (a) 90% (c) 97%
4 1
13. 890,000
4. (a) =
~
8 2
150 3
14. (a) =
~
2 1
200 4
(O s
4
17
(c)
26 1
5. (a) =
~
20
52 2
_85_ 17
=
48
= n 15. (a)
100 20
(c)
52
" 13
JJ2_ 23
(c) =
10 100 25
6. (a) -^
16. (a) J_
9 10
(c) tt
(c)
J_
7. (a) .4
10
(c) .7
8. (a)
j
Exercise Set 5.5

(page 303) 1. (a) The sample space consists of the (a) .42
four suits, {heart, club, diamond, (c) .4
or spade}, or the two colors, {red

(a) .35
or black}.
(c) .3
(c)
(e) 1
(e) .8
2 20
2. (a) = (g)
=
35
Chapter 5 523
11. .10 ,- , , 1
15. (a)
Chapter Test
(page 311) 1. l
2. 720
16
i ,
3. (a)
,
49
12 2
(b)
42
= 7
4. 35
S'
20 4
6. 48,450
Chapter 6
Exercise Set 6.1

Chapter 6 525
Exercise Set 6.2

(page 325) 1. (a) .6 (c) .4 (e) .28
Exercise Set 6.4

(page 343) 1. 60 17. 10.5
3. 5 19. (a) $187.50

(c) $2933.33
5. 2
21. 5
6. (a) 30
(c) 30 23. .24 (less than 1)
7. 3.5 25. $14.00 gain
9. (a) -2.74 cents 27. The scheme that costs $1 per ques-
tionnaire
11. 5.31 minutes
= 29. 11.5%
13. E(x) 3
15. .2 (less than 1)
Exercise 6.5
(page 354) 1. Is a probability matrix 25.
3. Is a probability matrix
5. Is not a probability matrix
7. Transition matrix
9. Not a transition matrix
16. (a) .3
17. (a) .4
(c) .4
19.
Si ,
5^ S3
20. (a) [.48 .52]
21. (a) [.452 .548]
22. (a) [ft ft]
23. (a) [ft ft"
ft ft_
24. (a) [^ W]
Chapter 6 527
54. (a)
181 system in state 2 will be in state

,c^
(g) Expected transitions = -rr- (t>2>-
4. is the expected number of
times the system in state 3 will
197
Expected transitions = -rr- (S3). be in state 2. fj is the expected
number of times the system in
be in state 3. I is the
state 3 will
Expected transitions = -r- (i 4 )-
expected number of times the
system in state 3 will be in state
(i) Expected transitions = y (i|).
4. i is the expected number of
Expected transitions = 25 (5 2 ).
be in state 2. ^ is the expected
Expected transitions
state 4 will be in state 3. I is the
6. (a) 1 is the expected number of expected number of times the

times the system in state 1 will system in state 4 will be in state
be in state 1. M is the expected 4.
number of times the system in (i) 5 is the expected number of

state 1 will be in state 3. if is the times the system in state 1 will
expected number of times the be in state 1 . is the expected

system in state 3 will be in state number of times the system in
1. H is the expected number of state 1 will be in state 2. f is the
times the system in state 3 will expected number of times the
be in state 3. system in state 1 will be in state
(c) I is the expected number of 4. 5 is the expected number of
times the system in state 1 will times the system in state 2 will
be in state 1. I is the expected be in state 1. 10 is the expected
number of times the system in number of times the system in
state 1 will be in state 3. I is the state 2 will be in state 2. 10 is the
expected number of times the expected number of times the

system in state 3 will be in state system in state 2 will be in state
1. V is the expected number of 4. 5 is the expected number of
times the system in state 3 will times the system in state 4 will
be in state 3. be in state 1. is the expected
(e) f is the expected number of number of times the system in
times the system in state 3 will state 4 will be in state 2. -T is the
be in state 3. is the expected expected number of times the

number of times the system in system in state 4 will be in state
state 3 will be in state 4. f is the 4.

1
system in state 4 will be in state
I 2
3. I isthe expected number of
times the system in state 4 will 9. (a) Yes
be in state 4.
(g) *? is the expected number of

be in state 2. ft is the expected
state 2 will be in state 3. I is the

Chapter 6 529
Exercise Set 6.7

(page 375) 1. (a) (i) Value = 0; fair; strategies =
"1"
(c) 3 in second row, first column; or
3 in first column, third row; or 3 [1 0] and
in second column, third row
(e) 4 in third row, third column (k) Value = 1; not fair; strategies
"0"
2. (a) 0; [1 0] and =
[:] [1 0] and
(c) 3; [0 1 0] and 47
4. (a) -TT-; not strictly determined
(c) t; not strictly determined

or [0 1] and
5. x = 0, 1, or 2
"0 12'
or [0 1] and -4-16
_@ 4
9. Yes .05
(e) 4; [0 1] and
A = [1,0] B =
1
3. (a) Value = 0; fair; strategies =

11. Democrats
III IV
[1 0] and I
J 300 100
Republicans
(c) Value = 0; fair; strategies = 400 150
Republicans [1 0]
[0 Hand I
J
= Democrats
(e) Value = 0; fair; strategies
[1 0] and Value = - 100
(g) Value = 0; fair; strategies =
'
[0 1] and
Exercise Set 6.8

(page 385) 44
1. (a) Value = tt; strategies = 2. (a) Value = -y; strategies =
11'
"0"
[A Aland
f 0]and
(c) Strictly determined; value = 3;
34
strategies = [1 0] and (c) Value = -68/- 14 = -=-; strate-
68
(e) Value = -p?; strategies = [A H]
15 gies = [0 H A] and
and
23 (c) Value = 9; strategies = [1 0]

3. (a) Value = t^; strategies =
and
[j]
[0 ft ft] and
(e) Value = -=-; strategies =
4 2
(c) Value = g = v strategies = [ft If] and
Jeannie
[| i 0] and I 2
Value = -1; strategies = 7. Kali.

(e)
T [1 0]
value = tj, strategies = [ft ft

and
and J
83
4. (a) Value = tt; strategies = 9. Value = 2; strategies = [1 0]
and
[fr A]and|J [J]
Value = 4; strategies = [1 0]
U. Democrats
(c)
Promise
and A B
Republicans 1 [2000 -1000
158
'jo
(e)
,,
Value =
.
-yr-; strategies
= Promise II [ 500 800
Value = ^j strategies
[A fflandlj*
[ft riland
5. (a) Value = jy: strategies =
[ft ft] and
Exercise Set 6.9

(page 391) (g) Reduced matrix
1. Value = =; strategies = [ft ft 0]
-2 5
and [*1 2 -3 -2
2
Value = =; strategies =
(c) Value = 1; strategies = [i j]
and
[I a] and
(e) Value = 2~'. strategies
2. (a) Strategies = [I f 0] and

[I I] and
Value =
Chapter 6 531
(c) Strategies = [0 0] and 5. They are identical.
7. Value = 2: strategies = [0 1]
0_
1 . 14
>. j-? to cotton and t= to peanuts
Value = ~
m - I
in + m - I n + m 11
Exercise Set 6.10

(page 398)
".565 .14 .295

9. (a) A 3
.45 .20 .35
.516 .096 .388
Therefore A is regular.
(c) [A
.323]
Chapter 7 533
4. (a)
f
35--
30--
25--
20-
15-
10 +
5--
n
5. (a) 24.5, 34.5, 44.5, 54.5, 64.5, 74.5,
84.5
(c)
7. (a)
140
60 60
30
n
c c
(O
-\ 1 1 1 1 1
Field
i 8 8 8 T? JS g
e a E
Chapter 7 535
(a) .267
10. (a)
/(thousands)
(C) Yes
13. 21.
Class Class Mark Class Class Marks
30-33 31.5 60-129 94.5

34-37 35.5 130-199 164.5
38-41 39.5 200-269 234.5
42-45 43.5 270-339 304.5
46-49 47.5 340-409 374.5
50-53 51.5 410-479 444.5
14.
Class Class Mark
43-50 46.5
51-58 54.5
59-66 62.5
67-74 70.5
75-82 78.5
83-90 86.5
91-98 94.5
IS. 50
17. 1978-1982 (100 cars)
19.
Units Relative
Sold Frequency
.02
.08
.12
24. (a) From labor, 45% was contrib-
uted to Democratic while only
.25
.26
3% to Republican candidates.
.19
From corporations, 16% was
contributed to Democratic while
.08
30% to Republican candidates.
1.00
(c) 5.94 million
Exercise Set 7.2

(page 427) 1. (a) Mean = 6 (c) Mean = 3.78 7. (a) Median (c) Mode
Median = 5 Median = 4 8. (a) 273.23; 275; 302
No mode Mode = 6
9. (a) 82.725 (c) 80
(e) Mean = 17.125
10. (a) 69.4
Median = 19.5
Mode = 23 11. (a) 2.78
3. 27 13. 2.375
4. (a) 2, 4, 6, 8 15. 99
(c) 20, 25, 25, 30 17. 118.21
(e) 1, 3,6, 6, 14
19. (a) The average salary of $15,200 is
5. 520 misleading since it is skewed on

Chapter 7 537
one end. (c) The median would 23. Mean = 71.5

be a better measurement.
25. 8.75 million gallons
21. (a) No. he did not use the fre-
quency.
Exercise Set 7.3

9. 296 15. 277.33 < m < 342.67

11. 0.0228 17. 9.85 < ti< 10.55
13. 236.08 < ix < 243.92
Review Exercises
(page 460) 1. (a) 18 (c) 20 (e) 8.11
3. x = 28.4, s = 12.48
4. (a) 75 (c) 94
5. (a) 20.18
6. (a)
fk
9. 101.5
10. (a) 77.25 <n < 82.75
11. (a) .67 (c) 1.44

7
13. E(x) = 5; p 128
7-13 14-20 21-27 28-34

15. 5; p = .9087; 1.55 standard devia-
tions below the mean
Chapter Test
(page 462) 1. 13
2. 13
3. 56
4./.
10--
1-4 5-8 9-12 13-16 17-20

Chapter 7 539
2.5 6.5 10.5 14.5 IX.5
6. 9.7
7. s = 4.86
8. 0.3413
9. 196.08 < fj.

< 203.92
10. 0.84
11. At least 75%
Index
Absolute value. 13 Binomial distribution, 334, 457

Absorbing Markov chain, 360 Break-even point, 89
Absorbing state, 360
Addition of matrices, 64
Addition property Cartesian coordinate system, 34
of equality, 19 Central limit theorem, 452
of inequality, 24 Chebyshev"s theorem, 437
Additive inverse of a matrix, 66 Class, 413
Amortization, 238 Class boundaries, 413
Amortized debt, 214 Class limits, 413
Annuity. 231 Class marks. 413
amount, 231 Closed model, 126
deferred, 246 Column matrix, 63
ordinary, 231 Combinations, 279
payment period of, 231 Common ratio, 227
present value of, 239 Commutative properties, 12
Area under normal curve. 444 Complements of sets. 10, 286
Arithmetic mean. 341, 422 Compound
Artificial variable, 199 amount, 216
Associative properties. 12 continuously, 225
Augmented matrix, 93 event, 260
Average. 341, 422 interest, 216
Conditional probability, 297
Confidence interval, 453
Bayes" probabilities, 321, 325 Confidence limit, 453
Bell-shaped curve, 442 Constant cost line, 153
Bernoulli. Jacob. 329 Constraints, 151

Bernoulli trials, 329 Continuous compounding, 225
541
542 Index
Continuous distribution, 457 Equivalent systems, 85

Coordinate, 1 Event(s), 260
axis. 34 complement, 286
plane, 34 compound, 260
system, 34 independent, 302
Corner point, 143 mutually exclusive, 288
Counting numbers, 1 probability of, 261
Counting techniques, 276 simple, 260
combinations, 279 Expectation, 337, 372
permutations, 276 Expected value, 338, 372, 426
tree diagram, 269 of a game, 372
Experiment, 257, 331
Dantzig, George, 141

Decision tree, 306
Factorial, 277
Demand curve. 28, 67
Fair game, 257, 339
Dependent events, 298
Feasible region
Depreciation, 43
bounded, 144
Dimension of a matrix, 63
Directed edge or graph, 136
corner point of, 143
Discrete probability function, 333

unbounded, 144
Finite, 332
Disjoint sets, 286
Finite stochastic processes, 314
Distribution
binomial, 457
First degree equation, 18
Fractions, 14
continuous, 457
frequency, 408 Frequency, 409
normal, 441, 457 Frequency distributions, 408, 409, 423

probability, 457
Frequency polygon, 414
Distributive property of multiplication over
Fundamental matrix, 364
addition, 12
Fundamental principle of counting, 270
Dominated rows and columns, 384 Future value (amount) of an annuity, 231
Dot product, 72
Dual problem, 191
Games
dominated, 384
Effective rate, 224 expected value, 372
Element fair, 339
of a matrix, 63 payoff, 401
of a set, 8 saddle point. 374
Empirical data, 267 strictly determined, 374
Empty set, definition, 9 strategy, 372, 381
Equal matrices, 64 value of game, 382, 383
Equal sets, 8 zero-sum, 371
Equations, 4, 18 Gauss- Jordan, 92, 96
equivalent, 19 Geometric progression, 227
graphs of, 34 Graph, 11, 34
linear, 18, 44 bar, 411
solutions of, 25 circle. 415
of value, 245 inequality, 39
Equilibrium line, 412
cost, 89 linear equations, 35
point, 37, 89, 351 Graphical solution, 82
revenue, 89 Greater than, 23
Equivalent solution sets, 25 Grouped frequency distribution. 408
Index 543
Half-planes, 38 Lines
Histogram, 413 horizontal, 51
Horizontal line, 49 parallel, 50
perpendicular, 50
vertical, 51
Identity, 12 Lines (equations of)
Identity matrix, 76 point-slope form, 81
Independent events, 301 slope intercept, 81
Indicators, 182
Inequalities, 4, 23
graph of, 39
greater than, 23 Markov chains, 347
less than, 23 absorbing, 360

properties of, 24 fixed matrix, 351
solution set, 25, 142 fundamental matrix, 364

Initial simplex tableau, 174 initial state, 350
Initial state. 350 probability matrix, 350
Input-output analysis. 130 regular, 352
Input-output matrix, 125 transition, 348
Integers, 11 steady-state. 351
Intercepts, 36 Matrix
Interest, 213 addition of, 64
compounded, 213, 216 augmented, 92
continuously compounded, 225 canonical form, 362
effective rate, 224 coefficient, 77
simple, 214 column, 63

Intersection of sets, 285 definition, 62
Interval notation, 24 dimension of, 63

Inventory matrix, 62 elements, 63
Inverse, 14, 114, 116, 120 equal, 64
identity, 76
inverse, 115
Leontief, Wassily, 125 Leontief input-output, 125

Leontief system, 125 negative. 66
closed, 126 product, 74
open, 128 dot, 72
Less than, 23 scalar, 66

Linear equation reduced, 108
definition, 18 row, 63, 94
graph of, 35 stochastic, 88
solution of, 25 sum, 64

Linear inequality, 24 square, 6
graphing, 38 transition, 348, 350
Linear model, 28 transpose of, 64

Linear programming. 159 Maximum, 156
duality, 191 Mean. 341, 422, 423

game theory, 393 Measure of central tendency, 422
graphical solution. 159 Median, 422, 424
maximum value. 160 Minimum. 156, 193
minimum value, 160 Mixed strategy, 371
nonstandard, 198 Mode. 422, 425
pivot element, 183 Multiplication rule, probabilities, 301
simplex, 181 Multiplicative inverse. 12
standard, 170 matrix, 115

544 Index
Natural numbers, 11 Probabilities {continued)

Negative of a matrix, 66 independent events, 301
Nominal rate, 217 mutually exclusive, 285
Nonbasic variables, 172 odds, 293
Normal distribution, 442 union rule, 290
Normal probability curve, 442 Probability density function, 333
Null set, 9 Probability matrix, 350
Problem constraints, 151
Product of
Objective function, 152 matrices, 74
Odds, 293 scalar and a matrix, 66
Open sentences, 4 Proper subset, 9
Optimal Pure strategy, 371
solution, 154
value, 151
Optimizing the function, 154 Quadrant, 34
Optimum strategy, 372, 381
Ordered pair, 34
Ordered triplets, 104 Random selection, 332
Origin, 11, 34 Random variable, 332
Outcome, 257 Range, 408

Outcome space, 257 Rates, nominal, 224
Rational numbers, 11
Real numbers, 10
Payment period, 239 Real number line, 1
Payoff matrix, 370 Reduced matrix, 108

Permissible row operations, 94 Region
Permutations, 276, 277 bounded, 144
Perpendicular, 50 feasible, 159
Perpetuity, 247 unbounded, 144
Pie chart, 415 Regular Markov chain, 352
Pivot Regular transition matrix, 353
column, 176 Relative frequency, 410
element, 175 Repeated trials, 329
operations, 176 Rise, 44
row, 183 Row matrix, 63
Point of equilibrium, 37 Row operation, 94
Point-slope form, 48 Run, 44
Points
break-even, 89
corner, 143 Saddle point, 374
Polya's four steps, 2 Sample, 452
Population, 452 Sample estimate, 452
Present value, 219, 239, 248 Sample space, 258
Principal, 213 uniform, 259
Probabilities, 256 Sampling distribution of the mean, 452
Bayes', 329 Scalar, 66
Bernoulli trials, 329 Set(s), 8
binomial, 331 complement of, 10
combinations, 279 disjoint, 286

complement, 292 element, 8
conditional, 297 empty or null, 9
distribution function, 333 equal, 8
empirical, 267 finite, 9
Index 545
Set(s) (continued) Sum of two matrices, 64

infinite, 9 Supply and demand, 28, 67
intersection of, 285 Surplus variable, 199
proper subset, 9 System of equations. 61
solution, 25 Systems of inequalities, 142
subset, 9 solution set for. 142
tabulation of, 8
union of, 285
universal, 9 Tableau, 174
Simple event, 260 Tabulation method, 8
Simple interest, 213 Transition matrix, 346
principal, 213 Transpose of a matrix, 64
Simplex method, 170 Tree diagram, 269
Simplex tableau, 174 Trial, 329
Sinking fund. 234 Two-person game, 370
Slack variable, 171
Slope. 45
Slope-intercept form. 48
Unbounded region, 144
Solution
Union of sets, 285
basic. 152
Uniform sample space, 259
Universe, 9
basic feasible, 172
feasible, 172
Solution of a maximum problem, 184
Value of a game, 281, 372
Solution of a minimum problem, 191
Variable
Solution of a system of equations, 120
artificial, 199
Solution set. 25
basic, 172
graph of, 35
nonbasic, 172
of a system of two equations, 82
random, 332
Square matrix, 63
slack, 171
Standard deviation, 432
surplus, 199
Standard maximization problem, 17
Variance, 432, 436
Standard units, 437
Venn diagram. 9
Statements, 4
Vertex, 134
Statistics, 407
Steady-state matrix, 351
Stochastic matrix, 88
Whole numbers, 11
Strategy of games
dominant rows (or columns), 384
linear programming, 393 x axis, 34
mixed, 371 .r-intercept, 36
optimum. 381
pure. 371
saddle point, 374 y axis, 34
strictly determined, 324 y-intercept, 36
two-by-two, 378
Strictly determined game, 324
Subset, 9 Zero matrix, 66
Subtraction of matrices, 67 Zero sum game. 371
Summation notation, 423 z-score, 437
Continued from inside front cover
Mazes, 6.6, 6.R Protein, 1.4
Medicine, 1.1, 3.6, 5.2, 5.3, 6.1, 6.2, 6.9, Psychological attraction and repulsion,
6.10, 7.R 2.3
Medicine (colds), 7.1 Psychology, 1.1, 6.6, 7.2
Military, 5.5, 6.1 Radar, 7.4
Mortality tables, 6.1, 6.4, 6.R
SAT scores, 1.4
Muscle efficiency, 1.2
Temperature, 1.4, 7.1
Nutrition, 2.3, 2.4, 3.3, 3.6
Testing techniques, 3.1, 6.3, 6.4
Nutrition (poultry), 3.R
Test scores, 1.4, 7.3, 7.4
Occupations, 7.1 Traffic flow, 2.5, 7.4
Transportation, 2.1, 2.2, 5.2, 7.1
Physical attributes (weight), 1.5, 7.2
Physical attributes (height), 1.5, 7.3, 7.4 Unemployment, 6.4
Plant nutrition, 3.3, 3.7
Vacation, 7.5
Politics, 1.2, 5.2, 5.4, 6.3, 6.4, 6.5, 6.7,
Vitamins, 7.2, 7.3
6.8, 7.1
Voting, 1.4, 2.3, 6.1
Polling, 2.6, 5.2, 5.5, 6.4
Pollution, 2.2, 2.5, 3.5, 7.2 Wildlife growth, 1.1
Population, 1.1, 1.5, 1.6, 2.3, 6.5, 7.3. Weights, 7.5
7.5
X-ray, 6.1
Population growth, 4.1
UCTOR'S COPY
Saunders College Publishing

ispleased to provide you with
this free examination copy
to aid in your
textbook selection process.
ISBN Q-DB-DSMSST-D

Finite Mathematics For Business and The Social and Life Sciences

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Finite Mathematics For Business and The Social and Life Sciences

Загружено:

Авторское право:

Доступные форматы

Finite Mathematics

for Business and the Social and Life Sciences [

FREE INSTRUCTOR'S COPY FREE INSTRUCTOR'S COPY J

BUSINESS AND ECONOMICS Equity, 4.4, 4.R

Bidding, 6.4 Graph theory, 2.R

Time analysis, 3.3

Inflation, 4.1, 4.3

Accidents, l.R, 5.2, 6.3, 6.4 Lottery, 6.4

forBusiness and the Social

Saunders College Publishing

Montreal Toronto London

Text Typeface: Times Roman

Printed in the United States of America

FINITEMATHEMATICS FOR BUSINESS AND THE SOCIAL AND LIFE SCIENCES:

Library of Congress Catalog Card Number: 90-053334

1234 039 987654321

THIS BOOK IS PRINTED ON ACID-FREE, RECYCLED PAPER

publication of this book

This book provides the noncalculus mathematics needed by students in man-

Chapter 1 (Problem Solving and Linear Models) is the foundation chapter

Chapter 6 (Additional Topics of Probability) discusses stochastic pro-

Chapter 7 (Statistics) introduces the basic concepts of descriptive sta-

A consistent pedagogical system is used in every chapter.

6000 = P(l + 0.02)

Steps in mathematical thinking are often repeated in nonmathematical terms

student will find an unusually largenumber of diagrams to assist in under-

EXERCISES Exercise sets can be either a strength or a weakness of a mathematics text-

An inexpensive scientific calculator or financial calculator should be pur-

important. The ability to group terms with parentheses is desirable. In addi-

ties is available to each class of students using this book. Undoubtedly,

Test Bank and Computerized Test Bank by Edmond Dixon of Tennessee

The following reviewers provided valuable suggestions for improving the

James Arnold, The University of Wisconsin at Milwaukee

My thanks to Professors David Foreman, Mary Hudson, and Fred

1 Problem Solving and Linear Models

1.1 Suggestions for Problem Solving 1

1.2 Foundations: Sets and the Real-Number

2.1 Getting Acquainted with Matrices 62

2.5 Systems with Three or More Variables or Three or More

3 Introduction to Linear Programming 141

3.1 Solving Systems of Inequalities Graphically 142

3.6 Minimization Using the Dual Problem 191

3.7 Mixed-Constraint Linear Programming 198

^4 Mathematics of Finance 212

4.1 Some Comparisons of Interest Rates 213

^5 Counting Techniques and Probability 256

5.1 The Language of Probability 257

6 Additional Topics of Probability 313

6.1 Finite Stochastic Processes 313

7.1 Frequency Distributions and Graphical Representations 408

1.1 Suggestions for Problem Solving

Four steps in problem solving

In the 1950s, a very successful research mathematician named George

Four Steps in Problem 1. Understand the problem.

In trying to "Understand the problem," we ordinarily ask a number of

3. Strip the problem of irrelevant details.

4. Don't impose conditions that do not exist.

Let's consider the first of these problems (a trick problem).

The need to strip the problem of irrelevant details is illustrated by these

Let's look at the first problem. Students often mistakenly concentrate

EXAMPLE 4 Translate the following into mathematical language:

(a) Six added to the product of 9 and a certain number

(a) 8 + 2 = 2 + 8 (a true statement of equality)