Decision Science

DECISION
SCIENCE
INDEX
003-059 --- MEASURE OF CENTRAL TENDENCY
060-118 --- PROBABLILITY 1
119-201 --- RANDOM VARIABLE
202-246 --- NORMAL VARIABLE 2
247-293 --- SAMPLING 2
294-336 --- CONFIDENCE INTERVAL 2
337-438 --- HYPOTHESIS TESTING 2
439-501 --- COMPARISON OF TWO VARIABLES
502-523 --- ANALYSIS OF VARIANCE
1-1
WHAT IS STATISTICS?
 Statistics is a science that helps us make better
decisions in business and economics as well as in
other fields.
 Statistics teaches us how to summarize, analyze,
and draw meaningful inferences from data that

then lead to improve decisions.
 These decisions that we make help us improve
the running, for example, a department, a

company, the entire economy, etc.
1-2
1-1. Using Statistics (Two Categories)

 Descriptive Statistics  Inferential Statistics
 Collect  Predict and forecast
 Organize values of
 Summarize population
parameters
 Display
 Test hypotheses about
 Analyze values of population
parameters
 Make decisions
1-3
Types of Data - Two Types

 Qualitative - Categorical  Quantitative - Measurable
or Nominal: or Countable:
Examples are- Examples are-
 Color  Temperatures
 Gender  Salaries
 Nationality  Number of points scored on

a 100 point exam
Nominal Level
• The lowest level of data measurement is the nominal. Numbers
representing nominal data can be used only to classify or categorize.
The numbers are used only to differentiate employees and not to
make a value statement about them.
Eg : sex, religion, ethnicity, geographic location, and place of birth
• Statistical techniques that are appropriate for analyzing nominal data
are limited. However, some of the more widely used statistics, such
as the chi-square statistic, can be applied to nominal data,
Ordinal Level
• Ordinal-level measurement can be used to rank or order objects. With
ordinal data, the distances or spacing represented by consecutive
numbers are not always equal.
Eg : a supervisor can evaluate three employees by ranking their
productivity with the numbers 1 through 3. However, the supervisor
could not establish that the intervals between the employees ranked
1 and 2 and between the employees ranked 2 and 3 are equal;
• Because nominal and ordinal data are often derived from imprecise
measurements such as demographic questions, the categorization of
people or objects, or the ranking of items, nominal and ordinal
data are nonmetric data and are referred to as qualitative data.
Interval Level
• The distances between consecutive numbers have meaning and the
data are always numerical. The distances represented by the
differences between consecutive numbers are equal; that is, interval
data have equal intervals.
• With interval-level data, the zero point is a matter of convention or
convenience and not a natural or fixed zero point. Zero is just another
point on the scale and does not mean the absence of the phenomenon
Eg : Temperature, percentage, change in employment, the
percentage return on a stock, and the dollar change in stock price.
Ratio Level
• Ratio data have the same properties as interval data, but ratio data
have an absolute zero, and the ratio of two numbers is
meaningful.
• The notion of absolute zero means that zero is fixed, and the zero
value in the data represents the absence of the characteristic being
studied. The value of zero cannot be arbitrarily assigned because
it represents a fixed point. This definition enables the statistician
to create ratios with the data.
Eg : height, weight, time, volume
Comparison of Four Levels of Data
• Statistical techniques can be separated into two categories:

parametric statistics and nonparametric statistics.
• Parametric statistics require that data be interval or ratio.
• If the data are nominal or ordinal, nonparametric
statistics
must be used
1-9
Samples and Populations
 A population consists of the set of all measurements

for which the investigator is interested.
 A sample is a subset of the measurements selected
from the population.
 A census is a complete enumeration of every item in
a population.
1-10
Simple Random Sample
 Sampling from the population is often done

randomly, such that every possible sample of
equal size (n) will have an equal chance of being
selected.
 A sample selected in this way is called a simple
random sample or just a random sample.

 A random sample allows chance to determine its
elements.
1-11
Samples and Populations
Population (N) Sample (n)

1-12
Why Sample?
⚫Census of a population may

be:
 Impossible
 Impractical
 Too costly
What is a measure of Central
Tendency?
• Type of measure that is used to describe a set of data yield

information about the center, or middle part, of a group
of numbers
• Measures of central tendency do not focus on the span of
the
data set or how far values are from the middle numbers
• mode, the median, the mean, percentiles, and quartiles.
1-15
Arithmetic Mean or
Average
◼ The arithmetic mean is the average of a group of numbers and is
computed by summing all numbers and dividing by the number of
numbers
◼ The population mean is represented by ( μ). The sample mean is

represented by (x)
1-16
Example
The magazine Forbes publishes annually a

list of the world’s wealthiest individuals.
For, 2007, the net worth of the 20 richest
individuals, in $billions, is as follows:
33,26,24,21,19,20,18,18,52,56,27,22,18,49,
22,20,23,32,20,18
Find mean.
1-17
Example – Mean
538
x
i
i
x  20  26.9
1
1-18
Percentiles
⚫ Percentiles are measures of central tendency that divide a group
of data into 100 parts
 The nth percentile is the value such that at least n percent of the
data are below that value and at most (100 - n) percent are above
that value
 Steps in Determining the Location of a Percentile
1. Organize the numbers into an ascending-order array.
2. Calculate the percentile location (P ) by (n+1)P /100 where, n is

the number of observations in the set.
1-19
Example - Billionaires
Billions Sorted Billions

33 18
26 18
24 18
21 18
19 19
20 20
18 20
18 20
52 21
56 22
27 22
22 23
18 24
49 26
22 27
20 32
23 33
32 49
20 52
18 56
1-20
Example - Percentiles
 Find the 50th, 80th and the 90th percentiles of this
data set.
 To find the 50th percentile, determine the data point
in position (n + 1)P/100 = (20 + 1)(50/100)
= 10.5.
 Thus, the percentile is located at the 10.5th
position.
 The 10th observation in the ordered set is 22, and
the 11th observation is also 22.
 The 50th percentile will lie halfway between the
10th and 11th values (which are both 22 in this case)
and is thus 22.
1-21
 To find the 80th percentile, determine the data

point in position (n + 1)P/100 = (20 + 1)(80/100)
= 16.8.
position.
 The 16th observation is 32, and the 17th
observation is also 33.
 The 80th percentile is a point lying 0.8 of the
way from 32 to 33 and is thus 32.8.
1-22
 To find the 90th percentile, determine the data

point in position (n + 1)P/100 = (20 + 1)
(90/100)
= 18.9.
position.
 The 18th observation is 49, and the 19th
observation is also 52.
 The 90th percentile is a point lying 0.9 of the
way from 49 to 52 and is thus 49 + 0.9(52 – 49)
= 49 + 0.93 = 49 + 2.7 = 51.7.
1-23
Quartiles – Special Percentiles
 Quartiles are the percentage points that

break down the ordered data set into
quarters.
 The first quartile is the 25th percentile.
It is the point below which lie 1/4 of the
data.
 The second quartile is the 50th percentile. It is
the point below which lie 1/2 of the data. This
is also called the median.
 The third quartile is the 75th percentile.
It is the point below which lie 3/4 of the
data.
1-24
Quartiles and Interquartile Range
 The first quartile, Q1, (25th percentile) is

often called the lower quartile.

The second quartile, Q2, (50th
percentile) is often called the median
or the middle quartile.

The third quartile, Q3, (75th percentile)
is often called the upper quartile.
 The interquartile range is the difference
between the first and the third quartiles.
1-25
Example : Finding Quartiles
Sorted (n+1)P/100 Quartiles

Billions Billions Position
33 18
26 18
24 18
21 18
19 19 First Quartile (20+1)25/100=5.25 19 + (.25)(1) = 19.25
20 20
18 20
18 20
52 21
56 22 Median (20+1)50/100=10.5 22 + (.5)(0) = 22
27 22
22 23
18 24
49 26
22 27 Third Quartile (20+1)75/100=15.75 27+ (.75)(5) = 30.75
20 32
23 33
32 49
20 52
18 56
Median
• The median is the middle value in an ordered array of
numbers
• For an array with an odd number of terms, the median
is the middle number – (n+1)/2th term
• For an array with an even number of terms, the
median is the average of the two middle numbers –
average of (n/2)th term and (n+1)/2th term.
1-27
Example – Median
Sorted
Billions
33
Billions 18
26
24
18
18
Median
21
19
18
19
50th Percentile
20 20
18 20
18 20 (20+1)50/100=10.5 22 +
52 21 (.5)(0) = 22
56 22
27 22
22 23 Median
18 24
49
22
26
27
The median is the middle
20
23
32
33 value of data sorted in
32 49
20 52 order of magnitude.
18 56
It is the 50th percentile.

Mode
• The mode is the most frequently occurring value in a set of data.

• In the case of a tie for the most frequently occurring value, two
modes are listed. Then the data are said to be bimodal
• Data sets with more than two modes are referred to as
multimodal.
• The mode is an appropriate measure of central tendency for
nominal-level data.
Mode - Examples
a. 5.40 1.10 0.42 0.73 0.48 1.10 Mode is 1.10
b. 27 27 27 55 55 55 88 88 99
Bimodal - 27 &
55
c. 1 2 3 6 7 8 9 10
No Mode
 Consider following data sets,
 Set I : 1,2,3,4,5,6,6,7,8,9,10,11 Mean ?

Median ?
 Set II : 4,5,5,5,6,6,6,6,7,7,7,8 Mode ?
Measures of variability
• To describe the spread or the dispersion of a set of data. Using
measures of variability in conjunction with measures of central
tendency makes possible a more complete numerical description of
the data.
1. Range - Difference between the largest value of a data set and the
smallest value of a set.
2. Interquartile Range - Range of values between the first and third
quartile. Essentially, it is the range of the middle 50% of the data
3. Variance - The average of the squared deviations about the
arithmetic mean for a set of numbers.
Computational Formulas for Variance
and Standard Deviation
1-33
Group Data
 Dividing data into groups or classes or intervals

 Groups should be:
 Mutually exclusive
 Not overlapping - every observation is assigned to only one group
 Exhaustive
 Every observation is assigned to a group
 Equal-width (if possible)
 First or last group may be open-ended
1-34
Frequency Distribution
 Table with two columns listing:

 Each and every group or class or interval of values
 Associated frequency of each group
 Number of observations assigned to each group
 Sum of frequencies is number of observations

 N for population
 n for sample
 Class midpoint is the middle value of a group or class or
interval
 Relative frequency is the percentage of total
observations
in each class
 Sum of relative frequencies = 1
1-35
Example : Frequency Distribution
x f(x) f(x)/n
Spending Class ($) Frequency (number of customers) Relative Frequency
0 to less than 100 30 0.163

100 to less than 200 38 0.207
200 to less than 300 50 0.272
300 to less than 400 31 0.168
400 to less than 500 22 0.120
500 to less than 600 13 0.070
184 1.000
• Example of relative frequency: 30/184 = 0.163

• Sum of relative frequencies = 1
1-36
Cumulative Frequency Distribution
x F(x) F(x)/n
Spending Class ($) Cumulative Frequency Cumulative Relative Frequency
0 to less than 100 30 0.163

100 to less than 200 68 0.370
200 to less than 300 118 0.641
300 to less than 400 149 0.810
400 to less than 500 171 0.929
500 to less than 600 184 1.000
The cumulative frequency of each group is the sum of the

frequencies of that and all preceding groups.
Mean – Grouped Data
Problem
Solution
Median – Grouped Data
Problem
Solution
Population Variance – Grouped Data
Problem
Solution
Solution
Sample Variance – Grouped Data
Calculate Mean, Median, Mode,
Variance & Standard Deviation
Class Interval Frequency

250-260 8
260-270 10
270-280 16
280-290 14
290-300 10
300-310 5
310-320 2
Calculate Mean, Median, Mode,
Variance & Standard Deviation
Class Interval Frequency
195-199 1
190-194 2
185-189 4
180-184 5
175-179 8
170-174 10
165-169 6
160-164 4
155-159 4
150-154 2
145-149 3
140-144 1
Measure of Shape
• Tools that can be used to describe the shape of a distribution of data.
1. Skewness - When a distribution is asymmetrical or lacks symmetry.
The skewed portion is the long, thin part of the curve

Measure of Shape
2. Kurtosis - It describes the amount of
peaked-ness of a distribution
2-1
2-1 Probability is:
 A quantitative measure of uncertainty

 A measure of the strength of belief in the occurrence of an
uncertain event
 A measure of the degree of chance or likelihood of
occurrence of an uncertain event
 Measured by a number between 0 and 1 (or between 0% and
100%)
2-2
Types of Probability - Classical

 When probabilities are assigned based on laws and rules, the
method is referred to as the classical method of assigning
probabilities. This method involves an experiment, which is a
process that produces outcomes, and an event, which is an
outcome of an experiment.
 The probability can be determined even before the part is sampled
because with the classical method, the probabilities can be
determined a priori; that is, they can be determined prior to the
experiment.
 examples: toss a coin, roll a die, pick a card
where
N = total possible number of outcomes of an experiment
n = the number of outcomes in which the event occurs out of N outcomes
e
2-3
Types of Probability - Subjective
 The subjective method of assigning probability is based on the

feelings or insights of the person determining the probability.
Subjective probability comes from the person’s intuition or
reasoning.
 example : An experienced airline mechanic can usually assign a
meaningful probability that a particular plane will have a certain type
of mechanical difficulty
2-4
2-2 Basic
Definitions
 Experiment
 It is a process that produces outcomes.

 Event
 It is an outcome of an experiment, the experiment

defines the possibilities of the event
 Sample Space
 Itis a complete roster or listing of all elementary
events for an experiment.
2-5
Basic Definitions (Continued)
 Intersection (And)
 A
– a set containing all elements in both A and
 B
Union (Or) B
 A
– a set containing all elements in A or B or
both B
2-6
Sets: A Intersecting with B
A
B
A
B
2-7
Sets: A Union B
A
B
A
B
2-8
• Mutually exclusive or disjoint sets

–sets having no elements in common, having no
intersection, whose intersection is the empty set
• Partition
–a collection of mutually exclusive sets which
together include all possible elements,
whose union is the universal set
2-9
Mutually Exclusive or Disjoint Sets
Sets have nothing in common
B
A
2-10

Independent Events
 Two or more events are independent events if the
occurrence or nonoccurrence of one of the events does not
affect the occurrence or nonoccurrence of the other
event(s).
 Rolling dice; each die is independent of the other.
Whether a 6 is rolled on the first die has no influence on
whether a 6 is rolled on the second die.
 Coin tosses always are independent of each other. The
event of getting a head on the first toss of a coin is
independent of getting a head on the second toss.
P (X/Y) = P (X) and P (Y/X) = P (Y)
2-11
2-3 Basic Rules for

Probability
⚫ Range of Values for P(A): 0  P(A)1
 Complements - Probability of not A
P( A)  1
P( A)
 Intersection - Probability of both A and B
A
P( A  B)  n(n(S )
 Mutually exclusiveB)
events (A and C) :
P( A  C )  0
2-12
Basic Rules for Probability

(Continued)
• Union - Probability of A or B or both (rule of
unions)
P( A  B)  n (n(S) A  B )  P( A)  P ( B )  P( A 
B) exclusive events: If A and B are mutually exclusive, then
 Mutually
P ( A  B )  0 so P ( A  B )  P ( A) 
P(B)
Sampling from a Population
with Replacement
Sampling n items from a population of size N with replacement
would provide
(N)n possibilities
where
N = population size
n = sample size
Q. Each time a die, which has six sides, is rolled, the outcomes are
independent (with replacement) of the previous roll. If a die is
rolled three times in succession, how many different outcomes
can occur?
Q. Each time a die, which has six sides, is rolled, the outcomes are
independent (with replacement) of the previous roll. If a die is
rolled three times in succession, how many different outcomes
can occur?
The size of the population, N, is 6, the six sides of the

die.
Three dice rolls, n = 3.
(N)n = (6)3 = 216
Combinations - Sampling from a
Population without
Replacement
Sampling n items from a population of size N with without
replacement would provide
where
N = population size
n = sample size
Q. Suppose a small law firm has 16 employees and three are to be
selected randomly to represent the company at the annual
meeting of the American Bar Association. How many different
combinations of lawyers could be sent to the meeting?
Q. Suppose a small law firm has 16 employees and three are to be
selected randomly to represent the company at the annual
meeting of the American Bar Association. How many different
combinations of lawyers could be sent to the meeting?
Types of Probabilities
Addition Laws
Q. Yankelovich Partners conducted a survey for the American
Society of Interior Designers in which workers were asked which
changes in office design would increase productivity. Respondents
were allowed to answer more than one type of design change. The
number one change that 70% of the workers said would increase
productivity was reducing noise. In second place was more
storage/filing space, selected by 67%. In addition, suppose 56% of
the respondents believed both noise reduction and increased
storage space would improve productivity. If one of the survey
respondents was randomly selected and asked what office design
changes would increase worker productivity, what is the probability
that this person would select reducing noise or more storage/filing
space?
Let N represent the event “reducing noise.”
S represent the event “more storage/ filing space.”
P (N U S) = P (N) + P (S) = .70 + .67 = 1.37

Let N represent the event “reducing noise.”
S represent the event “more storage/ filing space.”
P (N U S) = P (N) + P (S) = .70 + .67 = 1.37
P (N U S) = P (N) + P (S) - P (N ꓵ S) = .70 + .67 - .56 = .81

Q. The client company data from the Decision Dilemma
reveal that 155 employees worked one of four types of
positions. Shown here again is the raw values matrix (also
called a contingency table) with the frequency counts for
each category and for subtotals and totals containing a
breakdown of these employees by type of position and by
sex. If an employee of the company is selected
randomly, what is the probability that the employee is
female or a professional worker?
By the general law of
addition,
Q. Shown here are the raw values matrix and corresponding
probability matrix for the results of a national survey of 200 executives
who were asked to identify the geographic locale of their company and
their company’s industry type.
The executives were only allowed to select one locale and one
industry type.
Suppose a respondent is selected randomly from these data.
1 What is the probability that the respondent is from the Midwest (F)?
2 What is the probability that the respondent is from the
communications industry (C) or from the Northeast (D)?
3. What is the probability that the respondent is from the Southeast
(E) or from the finance industry (A)?
Multiplication Laws
Both event X and event Y will occur at the same time
Q. A manufacturing firm produces pads of bound paper.
Three percent of all paper pads produced are improperly
bound. An inspector randomly samples two pads of
paper, one at a time. Because a large number of pads are
being produced during the inspection, the sampling being
done, in essence, is with replacement. What is the
probability that the two pads selected are both improperly
bound?
Q. A manufacturing firm produces pads of bound paper.
Three percent of all paper pads produced are improperly
bound. An inspector randomly samples two pads of
paper, one at a time. Because a large number of pads are
being produced during the inspection, the sampling being
done, in essence, is with replacement. What is the
probability that the two pads selected are both improperly
bound?
Let I denote improperly bound

P (I1 ꓵ I2) = P (I1) P (I2) = (.03)(.03) = .0009
Q. According to the U.S. Bureau of Labor Statistics, 46% of the U.S.
labor force is female. In addition, 25% of the women in the labor
force work part time. What is the probability that a randomly
selected member of the U.S. labor force is a woman and works part-
time?
Q. According to the U.S. Bureau of Labor Statistics, 46% of the U.S.
labor force is female. In addition, 25% of the women in the labor
force work part time. What is the probability that a randomly
selected member of the U.S. labor force is a woman and works part-
time?
P (W ꓵ T) = P (W) . P (T / W)
P (W ꓵ T) = P (W) . P (T /W) = ( .46)( .

25) = .115
Q. A company has 140 employees, of which 30 are
supervisors. Eighty of the employees are married, and 20%
of the married employees are supervisors. If a company
employee is randomly selected, what is the probability
that the employee is married and is a supervisor?
Q. A company has 140 employees, of which 30 are
supervisors. Eighty of the employees are married, and 20%
of the married employees are supervisors. If a company
employee is randomly selected, what is the probability
that the employee is married and is a supervisor?
Let M denote married and S denote supervisor

P (M ꓵ S) = P (M) . P (S/M) = (.5714)(.20) = .
1143
2-35
2-4 Conditional Probability
• Conditional Probability - Probability of A given B
P ( A B )  P (PA( B) B ) , where P ( B ) 

0
To determine if X and Y are independent events, the following
must be true.
P( AB)  P( A)
P(B A)  P(B)
Q. In the study by Yankelovich Partners to determine what changes
in office design would improve productivity, 70% of the respondents
believed noise reduction would improve productivity and 67% said
increased storage space would improve productivity. In addition,
suppose 56% of the respondents believed both noise reduction and
increased storage space would improve productivity. What is the
probability that a randomly selected person believes storage space
would improve productivity given that he or she believes noise
reduction improves productivity?
Q. In the study by Yankelovich Partners to determine what changes
in office design would improve productivity, 70% of the respondents
believed noise reduction would improve productivity and 67% said
increased storage space would improve productivity. In addition,
suppose 56% of the respondents believed both noise reduction and
increased storage space would improve productivity. What is the
probability that a randomly selected person believes storage space
would improve productivity given that he or she believes noise
reduction improves productivity?
Q. Use these data to find:
Q. Use these data to find:
Q. Test the matrix for the 200
executive responses to
determine whether industry
type is independent of
geographic location.
Q. Test the matrix for the 200 executive responses
to determine whether industry type is
independent of geographic location.
Select one industry and one geographic location (say, A—Finance and G—West).
2-42
2-7 The Law of Total Probability and

Bayes’ Theorem
The law of total probability:
P( A)  P( A B)  P( A B )
In terms of conditional probabilities:

P( A)  P( A B)  P( A B )
 P( AB) P(B)  P( AB ) P(B )
More generally (where Bi make up a partition):

P( A)   P( A i
B )   P( ABi ) i
P(B )
Q. An analyst believes the stock market has a 0.75 probability of
going up in the next year if the economy should do well, and a 0.30
probability of going up if the economy should not do well during the
year. The analyst further believes there is a 0.80 probability that the
economy will do well in the coming year. What is the probability
that the stock market will go up next year ?
Q. An analyst believes the stock market has a 0.75 probability of
going up in the next year if the economy should do well, and a 0.30
probability of going up if the economy should not do well during the
year. The analyst further believes there is a 0.80 probability that the
economy will do well in the coming year. What is the probability
that the stock market will go up next year ?
Event U: Stock market will go up in the next year

Event W: Economy will do well in the next year
P ( U W ) .75
P(U W )  30
P ( W )  . 8 0  P ( W ) 1.8 .2
P(U )  P(U  W )  P(U  W )

 P(U W ) P(W)  P(U W ) P(W
)
 (.75)(.80)  (.30)(.20)
.60.06 .66
2-45
Bayes’
Theorem
• Bayes’ theorem enables you, knowing just a little more than the
probability of A given B, to find the probability of B given A.
• Based on the definition of conditional probability and the law of total
probability.
P( A  B)
P ( B A)  P ( A)
P( A  B) Applying the law of total
 probability to the denominator
P( A  B)  P( A 
B )
 Applying the definition of
P( A B ) P
P (( BA)B) P
P (( BA)B ) P ( B conditional probability throughout
)
Q. A company screens job applicants for illegal drug use at a certain stage
in their hiring process. The specific test they use has a false positive rate
of 2% and a false negative rate of 1%. Suppose that 5% of all applicants
are actually using illegal drugs, and we randomly select an applicant.
Given the applicant tests positive, what is the probability that they are
actually on drugs ?
actually on drugs ?
495
‘positive’
99%
500
On drugs 1%
5
5% ‘negative’
10,000 190
applicants ‘positive’
2%
9500
95% Not on drugs
98%
9
3
1
0
‘
n
e
g
a
actually on drugs ?
495
‘positive’
99%
500
On drugs 1%
5
5% ‘negative’
10,000 190
applicants ‘positive’
2%
9500
95% Not on drugs
98%
9
3
1
0
‘
n
e
g
a
Q. Consider a test for an illness. The test has a known reliability.
1. When administered to an ill person, the test will indicate so with
probability 0.92
2. When administered to a person who is not ill, the test will
erroneously give a positive result with probability 0.04
Suppose the illness is rare and is known to affect only 0.1% of the entire
population. If a person is randomly selected from the entire population
and is given the test and the result is positive, what is the posterior
probability that the person in ill ?
2-50
Example
P(I )  0.001 P(I Z) 

P(I  Z)
P(Z)
P(I  Z)

P(I  Z)  P(I  Z )
P(I )  0.999
P(Z I ) P(I )

P(Z I ) P(I )  P(Z I
) P(I )
P(Z I )  0.92  (.92)(0.001)
(.92)(0.001)  (0.04)(.999)
 0.00092
0.00092
0.0 0 0 9 2  0.0 3 9 9 6  .0 4 0 8 8
P(Z I )  0.04  .0 2 2 5
Q. A particular type of printer ribbon is produced by only two companies, Alamo
Ribbon Company and South Jersey Products. Suppose Alamo produces 65% of the
ribbons and that South Jersey produces 35%. Eight percent of the ribbons
produced by Alamo are defective and 12% of the South Jersey ribbons are
defective. A customer purchases a new ribbon. What is the probability that
Alamo produced the ribbon? What is the probability that South Jersey produced
the ribbon? The ribbon is tested, and it is defective. Now what is the probability
that Alamo produced the ribbon? That South Jersey produced the ribbon?
Q. Machines A, B, and C all produce the same two parts, X and Y. Of
all the parts produced, machine A produces 60%, machine B
produces 30%, and machine C produces 10%. In addition,
40% of the parts made by machine A are part X.
50% of the parts made by machine B are part
X. 70% of the parts made by machine C are
part X.
A part produced by this company is randomly sampled and is
determined to be an X part. With the knowledge that it is an X part,
revise the probabilities that the part came from machine A, B, or
C.
part X.
C.
part X.
C.
2-57
Q. An economist believes that during periods of high economic growth, the U.S.
dollar appreciates with probability 0.70; in periods of moderate economic
growth, the dollar appreciates with probability 0.40; and during periods of low
economic growth, the dollar appreciates with probability 0.20. During any period
of time, the probability of high economic growth is 0.30, the probability of
moderate economic growth is 0.50, and the probability of low economic growth
is 0.50. Suppose the dollar has been appreciating during the present period.
What is the probability we are experiencing a period of high economic growth?
2-58
Partition: Event A  Appreciation

H - High growth P(H) = 0.30 P( A H)  0.70
M - Moderate growth P(M) = 0.50 P( A M ) 
L - Low growth P(L) = 0.20 0.40 P( A L) 
0.20
P(H  A)
P(H A) 
P(H 
 P( A)
P(H  A) A) P( M  A)  P( L 
A)
 P( A H) P(H)
P( A H) P(H)  P( A M ) P( M )
  P( A L) P( L)
(0.70)(0.30)

(0.70)(0.30)
0.210.20  (0.40)(0.50)  (0.20)
 0.04
 0.467
0.45 0.21 (0.20)
0.21

2-59
Tree Diagram
Prior Conditional Joint

Probabilities Probabilities Probabilities
P( A H)  P( A  H) 
0.70 (0.30)(0.70) 
0.21
P( A H)  0.30
P ( H )  0.30 P( A  H)  (0.30)(0.30)  0.09
P( A M )  0.40 P( A  M )  (0.50)(0.40)  0.20
P( M )  0.50
P( A M )  0.60 P( A  M )  (0.50)(0.60)  0.30

P( A L)  0.20
P( L)  0.20 P( A  L)  (0.20)(0.20)  0.04
P( A L)  0.80 P( A  L)  (0.20)(0.80)  0.16

Random Variable
Example
Consider the experiment of tossing two six-sided dice. There are 36
possible
outcomes.
the two dice:
Let the random variable X represent the sum of the numbers on
In statistical experiments involving chance, outcomes occur
randomly
x P(x)* Probability Distribution of Sum of Two Dice
2 1/36
3 2/36 0.1
7
2 3 4 5 6 7 4 3/36
5 4/36
1,1 1,2 1,3 1,4 1,5 1,6 8 6 5/36
0.1
2
p(x
2,1 2,2 2,3 2,4 2,5 2,6 9 7 6/36
)
0.0
3,1 3,2 3,3 3,4 3,5 3,6 10 8 5/36 7
4,1 4,2 4,3 4,4 4,5 4,6 11 9 4/36

10 3/36 0.0
5,1 5,2 5,3 5,4 5,5 5,6 12 11 2/36
2 2 3 4 5 6 7 8 9 10 11 12
x
6,1 6,2 6,3 6,4 6,5 6,6 12 1/36
1
* Note that: P(x)  (6  (7  x)2 ) /
36
A random variable, say X, is a an uncertain quantity whose value depends
on chance.
3-3
Discrete and Continuous Random

Variables
A discrete random variable:
 has a countable number of possible values
 has discrete jumps (or gaps) between successive values
 has measurable probability associated with individual values
 counts
A continuous random variable:

 has an uncountably infinite number of possible values
 values are measured and not counted
 moves continuously from value to value
 has no measurable probability associated with each value
 measures (e.g.: height, weight, speed, value, duration,
length) data are measured and recorded, they become discrete data
Once continuous
because the data are rounded off to a discrete number
3-4
Rules of Discrete Probability

Distributions
The probability distribution of a discrete random

variable X must satisfy the following two conditions.
1. P(x)  0 for all values of
x. 2. P(x)  1
all x
Corollary: 0  P(X)  1
Problem
800, 900 and Now : the 500 Telephone
Numbers
The new code 500 is for busy, affluent people who
travel a lot; It can work with a cellular phone, your
home phone, office phone, second-home phone,
x P(x)
up to five additional phones besides your regular
0 0.1
one. The computer technology behind this service 1 0.2
is astounding – it first ring you up at the 2 0.3
telephone number specified as primary number. If 3 0.2
there is no answer, the computer switches to 4 0.1
search for you at your second-specified phone 5 0.1
number; if you do not answer there, it will switch 1.00
to your third phone; and so on up to five
allowable switches. From the data available on
an experimental run, the following probability
distribution is constructed for the number of
dialing switches that are necessary before a
3-6
Cumulative Distribution Function
The cumulative distribution function, F(x), of a discrete

random variable X is:
F(x)  P(X  x)  P(i)

all ix
Cumulative Probability Distribution of the Number of Switches

x P(x) F(x)
1 .0
0 0.1 0.1 0 .9
1 0.2 0.3 0 .8
0 .7
2 0.3 0.6 0 .6
F(x)
0 .5
3 0.2 0.8 0 .4
0 .3
4 0.1 0.9 0 .2
5 0.1 1.0 0 .1
0 .0
1.00 0 1 2
x
3 4 5
It lists the probabilities of occurrence of different values of random

3-7
The probability that at most three switches will occur:
x P(x) F(x)
0 0.1 0.1
1 0.2 0.3
2 0.3 0.6
3 0.2 0.8
4 0.1 0.9
5 0.1 1.0
1
Note: P(X < 3) = F(3) = 0.8 = P(0) + P(1) + P(2) + P(3)

3-8
Using Cumulative Probability

Distributions
The probability that more than one switch will occur:
x P(x) F(x)
0.1 0.1
0
1 0.2 0.3
2 0.3 0.6
3 0.2 0.8
4 0.1 0.9
5 0.1 1.0
1
Note: P(X > 1) = P(X > 2) = 1 – P(X < 1) = 1 – F(1) = 1 – 0.3 = 0.7
3-9
Using Cumulative Probability

Distributions
The probability that anywhere from one to three

switches will occur:
x P(x) F(x)
0 0.1 0.1
1 0.2 0.3
2 0.3 0.6
3 0.2 0.8
4 0.1 0.9
5 0.1 1.0
1
Note: P(1 < X < 3) = P(X < 3) – P(X < 0) = F(3) – F(0) = 0.8 – 0.1 = 0.7
3-10
Expected Values of Discrete

Random Variables
The mean of a probability distribution is a
measure of its centrality or location, as is the
mean or average of a frequency distribution. It is
a weighted average, with the values of the
random variable weighted by their probabilities. 0 1 2
2.3
3 4 5
The mean is also known as the expected value (or expectation) of a

random variable, because it is the value that is expected to occur, on
average. x P(x) xP(x)
0 0.1 0.0
The expected value of a discrete random 1 0.2 0.2
variable X is equal to the sum of each 2 0.3 0.6
value of the random variable multiplied 3 0.2 0.6
by its probability. 4 0.1 0.4
5 0.1 0.5
  E ( X )   xP(x) 1.0 2.3 = E(X) = 
all x
3-11
A Fair Game
Suppose you are playing a coin toss game in which you are
paid $1 if the coin turns up heads and you lose $1 when the
coin turns up tails. The expected value of this game is E(X)
=
0. A game of chance with an expected payoff of 0 is called a
fair game.
x P(x) xP(x)
-1 0.5 -0.50
1 0.5 0.50 -1
0
1
1.0 0.00 = E(X)=

3-12
Expected Value of a Function of a

Discrete Random Variables
The expected value of a function of a discrete random variable X is:
E[h( X )]   h(x) P(x)

all x
Example: Monthly sales of a certain product Number

are believed to follow the given probability of items, x P(x) xP(x)
5000 0.2 1000
distribution. Suppose the company has a fixed
6000 0.3 1800
monthly production cost of $8000 and that each 7000 0.2 1400
item brings $2. Find the expected monthly 8000 0.2 1600
profit h(X), from product sales. 9000 0.1 900
1.0 6700
h (X) = 2X – 8000 where X = # of items sold

3-13
Expected Value of a Function of a

Discrete Random Variables
The expected value of a function of a discrete random variable X is:
E[h( X )]   h(x) P(x)

all x
Example: Monthly sales of a certain Number

product are believed to follow the given of items, x P(x) xP(x) h(x)
h(x)P(x)
probability distribution. Suppose the 5000 0.2 1000 2000 400
company has a fixed monthly production 6000 0.3 1800 4000 1200
cost of $8000 and that each item brings 7000 0.2 1400 6000 1200
$2. Find the expected monthly profit 8000 0.2 1600 8000 1600
9000 0.1 900 10000 1000
h(X), from product sales.
E[h( X )]   h(x) P(x)  5400
1.0 6700 5400
all x Note: h (X) = 2X – 8000 where X = # of items sold
The expected value of a linear function of a random variable is:

E(aX+b)=aE(X)+b
In this case: E(2X-8000)=2E(X)-8000=(2)(6700)-8000=5400
3-14
Variance and Standard Deviation of

a Random Variable
The variance of a random variable is the expected

squared deviation from the mean:
2
  V ( X )  E [( X   ) 2 ]   all( xx   ) 2 P ( x )
   
2
 E ( X 2 )  [ E ( X )] 2    x 2 P ( x )    xP ( x )
 all x  all x

The standard deviation of a random variable is the

square root of its variance:   SD(X)  V(X)
Population Variance
Solution
a Random Variable
x P(x)
0 0.1
1 0.2
2 0.3
3 0.2
4 0.1
5 0.1
1.00
a Random Variable
2  V( X)  E[(X  )2]
Number of   (x  )2P(x) 
Switches, x P(x) xP(x) (x-) (x-)2 (x-)2P(x) x2P(x)
0 0.1 0.0 -2.3 5.29 0.529 0.0
2.01
1 0.2 0.2 -1.3 1.69 0.338 0.2 all x
2 0.3 0.6 -0.3 0.09 0.027 1.2
3 0.2 0.6 0.7 0.49 0.098 1.8
4 0.1 0.4 1.7 2.89 0.289 1.6  E( X 2 )
5 0.1 0.5 2.7 7.29 0.729 2.5
2.3 2.010 7.3  X)]2  
 [E( 2

 = 2.3.   x2P(x)   xP(x)
Recall: 
all x
 all 
x
 7.3 2.32 
2.01
Variance of a Linear Function of a
Random Variable
The variance of a linear function of a random variable is:
V(a X b) a2V(X) a22
 V(X)
2
Number  E( X 2 ) 
of items, x P(x) xP(x) x P(x)
2
[E(
 X )]2   
2
5000 0.2 1000 5000000
   x 2 P(x)    xP(x)
6000 0.3 1800 10800000 all x all x
7000 0.2 1400 9800000  46500000  (67002 )  1610000
8000 0.2 1600 12800000
9000 0.1 900 8100000   SD( X ) 
1.0 6700 1610000  1268.86
46500000 V (2 X  8000)(4)(1610000)
(2 2 )V ( X )  6440000
(2x  SD(2x  8000)
8000)
 2 x  (2)(1268.86) 
2537.72
Problem
During one holiday season, the Texas lottery played a game
called the Stocking Stuffer. With this game, total instant
winnings of $34.8 million were available in 70 million $1
tickets, with ticket prizes ranging from $1 to $1,000. Shown
here are the various prizes and the probability of winning
each prize. Use these data to compute the expected value
of the game, the variance of the game, and the standard
deviation of the game.
Solution
Solution
Sum and Linear Composites of
Random Variables
The mean or expected value of the sum of random
variables is the sum of their means or expected values:
( XY)  E(X Y)  E(X)E(Y) X  Y
 and E(Y) = $200
For example: E(X) =$350
E(X+Y) = $350 + $200 = $550
The variance of the sum of mutually independent random
variables is the sum of their variances:
 2 ( X Y)  V( X  Y)  V ( X ) V (Y)   2 X   2 Y
if and only if X and Y are independent.
For example: V(X) = 84 and V(Y) = 60 V(X+Y) =

144
Sum Variances (Additional)
Random Variables (Continued)
NOTE: E(X1 X 2 ... X k ) E(X1) E(X 2 )...
E(X k )
E(a1X1a2X2...ak Xk)a1E(X1)a2E(X2)...ak E(X k )
The variance of the sum of k mutually independent random

variables is the sum of their variances:
V (X1 X2 ... X k )V (X1)V (X2 )...V (X k )
and
V (a1X1a2X2...ak Xk )a2 V (X1)a2V (X2)...a2V (X k )
1
Random Variables (Continued)
A portfolio includes stocks in three
industries: financial, energy, and
The mean of the sum of the three
consumer goods. Assume that the three
sectors are independent of each other. random variables is: 1,000 +
The expected annual return and 1,200
standard deviations are as follows: + 600 = $2,800.
Financial – 1,000, and 700; energy –
1,200 and
1,100; consumer goods – 600 and 300.
What is the mean and standard deviation
of the annual return on this portfolio?
The variance of the three random variables for the three

sectors, assuming independence is: 7002 + 1,1002 + 3002 =
1,790,000.
Thus the standard deviation is (1,790,000) = $1,337.9.
Standard Distributions
Two discrete distributions:
1. binomial distribution
2. Poisson distribution
Bernoulli Random Variable
• If an experiment consists of a single trial and the outcome of

the trial can only be either a success* or a failure, then the trial
is called a Bernoulli trial.
• The number of success X in one Bernoulli trial, which can be 1
or 0, is a Bernoulli random variable.
• Note: If p is the probability of success in a Bernoulli
experiment, then P(1) = p, P(0) = 1 – p, E(X) = p and V(X) =
p(1 – p).
* Theterms success and failure are simply statistical terms, and do not have
positive or negative implications. In a production setting, finding a defective
product may be termed a “success,” although it is not a positive result.
The Binomial Random Variable
Consider a Bernoulli Process in which we have a sequence of n

identical trials satisfying the following conditions:
1. Each trial has two possible outcomes, called success *and failure.
The two outcomes are mutually exclusive and exhaustive.
2. The probability of success, denoted by p, remains constant from
trial to trial. The probability of failure is denoted by q, where q = 1-
p.
3. The n trials are independent. That is, the outcome of any trial does
not affect the outcomes of the other trials.
A random variable, X, that counts the number of successes in n Bernoulli
trials, where p is the probability of success* in any given trial, is said to
follow the binomial probability distribution with parameters n
(number of trials) and p (probability of success). We call X the binomial
random variable.
* The terms success and failure are simply statistical terms, and do not have positive or negative implications. In a
production setting, finding a defective product may be termed a “success,” although it is not a positive result.
3-30
Binomial Probabilities (Introduction)

Suppose we toss a single fair and balanced coin five times in
succession, and let X represent the number of heads.
There are 25 = 32 possible sequences of H and T (S and F) in the sample

space for this experiment. Of these, there are 10 in which there are exactly 2
heads (X=2):
HHTTT HTHTH HTTHT HTTTH THHTT THTHT THTTH TTHHT

TTHTH TTTHH
The probability of each of these 10 outcomes is p3q3 = (1/2)3(1/2)2=(1/32), so the

probability of 2 heads in 5 tosses of a fair and balanced coin is:
P(X = 2) = 10 * (1/32) = (10/32) = 0.3125

3-31
Binomial Probabilities (continued)
P(X=2) = 10 * (1/32) = (10/32) = .3125

Notice that this probability has two parts:
10 (1/32)
Number of outcomes Probability of each
with 2 heads outcome with 2 heads
In general:
1. The probability of a given sequence 2. The number of different sequences of n trials that
of x successes out of n trials with result in exactly x successes is equal to the number
probability of success p and of choices of x elements out of a total of n elements.
probability of failure q is equal to: This number is denoted:
 n n!
pxq(n-x) nCx    
 x  x!(n  x)!
3-32
The Binomial Probability Distribution
The binomial probability distribution: N um ber of

successes, x P robability
P(x n) !
n! x! 0
 n x ( n x)
0 (n0)
0 !( n  p q
P(x)    p q  (n  x)! px q ( n x) 0n )! !
 x 1 1 (n1)
where : 1!( n  p q
p is the probability of success in a single trial, 2 1)n !!
2 !( n  p 2q (n2)
q = 1-p, 2)! n!
n is the number of trials, and 3 3 (n3)
3 !( n  p q
x is the number of successes.  3) !

n n! p nq (nn)
n !( n 
n1) .!
00
Problem
For a binomial distribution with n = 7 and p = 0.2,
find
(a) P(r = 5).
(b) P(r > 2).
(c) P(r < 8).
(d) P(r > 4).
Problem
Problem
A Gallup survey found that 65% of all financial consumers
were very satisfied with their primary financial institution.
Suppose that 25 financial consumers are sampled and if the
Gallup survey result still holds true today, what is the
probability that exactly 19 are very satisfied with their primary
financial institution?
Problem
A Gallup survey found that 65% of all financial consumers
were very satisfied with their primary financial institution.
Suppose that 25 financial consumers are sampled and if the
Gallup survey result still holds true today, what is the
probability that exactly 19 are very satisfied with their primary
financial institution?
Problem
According to the U.S. Census Bureau, approximately 6% of all
workers in Jackson, Mississippi, are unemployed. In
conducting a random telephone survey in Jackson, what is the
probability of getting two or fewer unemployed workers in a
sample of 20?
Problem
According to the U.S. Census Bureau, approximately 6% of all
workers in Jackson, Mississippi, are unemployed. In
conducting a random telephone survey in Jackson, what is the
probability of getting two or fewer unemployed workers in a
sample of 20?
3-40
Using the Binomial Table
n=5
p
x 0.01 0.05 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.95 0.99
0 .951 .774 .590 .328 .168 .078 .031 .010 .002 .000 .000 .000 .000
1 .999 .977 .919 .737 .528 .337 .187 .087 .031 .007 .000 .000 .000
2 1.000 .999 .991 .942 .837 .683 .500 .317 .163 .058 .009 .001 .000
3 1.000 1.000 1.000 .993 .969 .913 .813 .663 .472 .263 .081 .023 .001
4 1.000 1.000 1.000 1.000 .998 .990 .969 .922 .832 .672 .410 .226 .049
h F(h) P(h)
Cumulative Binomial Deriving Individual Probabilities
Probability Distribution and 0 0.031 0.031
from Cumulative Probabilities
Binomial Probability 1 0.187 0.156
Distribution of H,the 2 0.500 0.313 F (x)  P( X  x)   P(i)
Number of Heads 3 0.813 0.313 all i x
Appearing in Five Tosses of 4 0.969 0.156

P(X) = F(x) - F(x - 1)
a Fair Coin 5 1.000 0.031 For example:
1.000
P(3)  F (3)  F (2)
.813.500
.313
3-41
Calculating Binomial Probabilities

60% of Brooke shares are owned by LeBow. A random
sample of 15 shares is chosen. What is the
probability that at most three of them will be found to be
owned by LeBow?
3-42
Problem
60% of Brooke shares are owned by LeBow. A random

sample of 15 shares is chosen. What is the
probability that at most three of them will be found to be
owned by LeBow?
n=15 p
.50 .60 .70
0
1
.000
.000
.000
.000
.000
.000

F (x)  P( X  x) alli P(i)
x
2 .004 .000 .000
F (3)  P( X  3) 
3 .018 .002 .000 0.002
4 .059 .009 .001
... ... ... ...
Problem
According to Information Resources, which publishes data
on market share for various products, Oreos control about
10% of the market for cookie brands. Suppose 20 purchasers
of cookies are selected randomly from the population. What
is the probability that fewer than four purchasers choose
Oreos?
Problem
According to Information Resources, which publishes data
on market share for various products, Oreos control about
10% of the market for cookie brands. Suppose 20
purchasers of cookies are selected randomly from the
population. What is the probability that fewer than four
purchasers choose Oreos?
Problem
A manufacturing company produces 10,000 plastic mugs per
week. This company supplies mugs to another company, which
packages the mugs as part of picnic sets. The second
company randomly samples 10 mugs sent from the supplier. If
two or fewer of the sampled mugs are defective, the second
company accepts the lot. What is the probability that the lot will
be accepted if the mug manufacturing company actually is
producing mugs that are 10% defective? 20% defective? 30%
defective? 40% defective?
Problem
3-47
Mean, Variance, and Standard

Deviation of the Binomial
Distribution
Mean of a binomial distribution:
For example, if H counts the number
  E ( X )  np of heads in five tosses of a fair coin :
Variance of a binomial distribution:   E(H )  (5)(.5)  2.5

H
  V ( X )  npq
2
  V (H )  (5)(.5)(.5)  1.25
H
2
Standard deviation of a binomial distribution:

  SD(H )  1.25  1.118
H
 = SD(X) = npq
3-48
Shape of the Binomial Distribution

p = 0.1 p = 0.3 p = 0.5
Binomial Probability: n=4 p=0.1 Binomial Probability: n=4 p=0.3 Binomial Probability: n=4 p=0.5
0.7 0.7 0.7
0.6 0.6 0.6
0.5 0.5 0.5
n=4 0.4 0.4 0.4

P(x
P(x
P(x
)
)
0.3 0.3 0.3
0.2 0.2 0.2
0.1 0.1 0.1
0.0 0.0 0.0

0 1 2 3 4 0 1 2 3 4 0 1 2 3 4
x x x
Binomial Probability: n=10 p=0.1 Binomial Probability: n=10 p=0.3 Binomial Probabil i ty: n=10 p=0.5
0. 5 0. 5 0. 5
0. 4 0. 4 0. 4
n = 10 0. 3 0. 3 0. 3
P(x
P(x
P(x
)
)
)
0. 2 0. 2 0. 2
0. 1 0. 1 0. 1
0. 0 0. 0 0. 0
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
x
x x
0.2 0.2 0.2
n = 20
P(x
P(x
P(x
)
)
0.1 0.1 0.1
0.0 0.0 0.0

0 1 2 3 4 5 6 7 8 9 1011121314151617181920 0 1 2 3 4 5 6 7 8 9 1011121314151617181920 0 1 2 3 4 5 6 7 8 9 1011121314151617181920
x x x
Binomial distributions become more symmetric as n increases and as p 0.5.

Problem
Harley Davidson, director of quality control for the Kyoto Motor
company, is conducting his monthly spot check of automatic
transmissions. In this procedure, 10 transmissions are
removed from the pool of components and are checked for
manufacturing defects. Historically, only 2 percent of the
transmissions have such flaws. (Assume that flaws occur
independently in different transmissions.)
(a) What is the probability that Harley’s sample contains more
than two transmissions with manufacturing flaws?
(b) What is the probability that none of the selected
transmissions has any manufacturing flaws?
Solution
Problem
Diane Bruns is the mayor of a large city. Lately, she has
become concerned about the possibility that large numbers of
people who are drawing unemployment checks are secretly
employed. Her assistants estimate that 40 percent of
unemployment beneficiaries fall into this category, but Ms.
Bruns is not convinced. She asks one of her aides to conduct
a quiet investigation of 10 randomly selected unemployment
beneficiaries.
(a)If the mayor’s assistants are correct, what is the
probability that more than eight of the individuals investigated
have jobs? (Do not use the tables.)
(b)If the mayor’s assistants are correct, what is the probability
that three of the investigated individuals have jobs? (Do not
use the tables.)
Solution
Poisson Distribution
• The Poisson distribution focuses only on the number of discrete
occurrences over some interval or continuum (number of random arrivals
per some time interval)
• A Poisson experiment does not have a given number of trials (n) as a
binomial
experiment.
Example, whereas a binomial experiment might be used to determine
how many U.S.- made cars are in a random sample of 20 cars, a
Poisson experiment might focus on the number of cars randomly
arriving at an automobile repair facility during a 10-minute interval.
• The Poisson distribution describes the occurrence of rare events. It is
referred to
as the law of improbable events.
• If the number of arrivals per interval is too frequent, the time interval can
be reduced enough so that a rare number of occurrences is expected.
Poisson Distribution - Characteristics
• It is a discrete distribution.
• It describes rare events.
• Each occurrence is independent of the other occurrences.
• It describes discrete occurrences over a continuum or interval.
• The occurrences in each interval can range from zero to infinity.
• The expected number of occurrences must hold constant
throughout the experiment
Examples of Poisson-type situations include the following:

1. Number of telephone calls per minute at a small business
2. Number of hazardous waste sites per county in the United States
3. Number of sewing flaws per pair of jeans during production
4. Number of times a tire blows on a commercial airplane per week
Poisson Formula
(the number of occurrences per interval)
The λ value must hold constant throughout a Poisson experiment.
Example:
• The average number of customers arriving at a Big-Bazaar store during a one-minute interval
will vary from hour to hour, day to day, and month to month.
• The number of flaws per pair of jeans might vary from Monday to Friday.
Problem
Suppose bank customers arrive randomly on weekday afternoons at
an average of 3.2 customers every 4 minutes. What is the probability
of exactly 5 customers arriving in a 4-minute interval on a weekday
afternoon?
Problem
Suppose bank customers arrive randomly on weekday afternoons at
an average of 3.2 customers every 4 minutes. What is the probability
of exactly 5 customers arriving in a 4-minute interval on a weekday
afternoon?
Problem
Bank customers arrive randomly on weekday afternoons at an
average of 3.2 customers every 4 minutes. What is the
probability of having more than 7 customers in a 4-minute
interval on a weekday afternoon?
Problem
Bank customers arrive randomly on weekday afternoons at an
average of 3.2 customers every 4 minutes. What is the
probability of having more than 7 customers in a 4-minute
interval on a weekday afternoon?
Problem
A bank has an average random arrival rate of 3.2
customers every 4 minutes. What is the probability of
getting exactly 10 customers during an 8-minute
interval?
Problem
A bank has an average random arrival rate of 3.2 customers
every 4 minutes. What is the probability of getting exactly 10
customers during an 8-minute interval?
Sol.
Never adjust or change x in a problem. Just because 10 customers arrive
in one 8-minute interval does not mean that there would necessarily
have been five customers in a 4-minute interval. There is no guarantee
how the 10 customers are spread over the 8-minute interval. Always
adjust the lambda value.
Using the Poisson Tables
Problem
If a real estate office sells 1.6 houses on an average weekday
and sales of houses on weekdays are Poisson distributed,
what is the probability of selling exactly 4 houses in one day?
What is the probability of selling no houses in one day? What
is the probability of selling more than five houses in a day?
What is the probability of selling 10 or more houses in a day?
What is the probability of selling exactly 4 houses in two
days?
Mean and Standard Deviation of a
Poisson Distribution
• The mean or expected value of a Poisson distribution is λ

(mean of a Poisson distribution gives a feel for the actual
occurrences that are likely to happen)
• The variance of a Poisson distribution also is λ

Approximating Binomial Problems
by the Poisson Distribution
• Binomial problems with large sample sizes and small values of p, which then
generate rare events can be approximated by Poisson
• if n > 20 and n.p < 7, the approximation is close enough to use the
Poisson distribution for binomial problems.
• Mean of the binomial distribution, μ = n. p , because μ is the expected
value of the binomial, it translates to the expected value, λ, of the Poisson
distribution. Using μ as the λ value and using the x value of the binomial
problem allows approximation of the probability from a Poisson table or
by the Poisson formula.
Problem
Q. n = 50 and p = .03. What is the probability that x = 4?
Problem
Q. n = 50 and p = .03. What is the probability that x = 4?
3-68
Continuous Random Variables
A continuous random variable is a random variable that can take on any value in
an interval of numbers.
The probabilities associated with a continuous random variable X are determined by the
probability density function of the random variable. The function, denoted f(x), has the
following properties.
1. f(x)  0 for all x.

2. The probability that X will be between two numbers a and b is equal to the
area
3. under f(x) between a and b.
The total area under the curve of f(x) is equal to 1.00.
The cumulative distribution function of a continuous random variable:
F(x) = P(X  x) =Area under f(x) between the smallest possible value of X (often -)
and the point x.
3-69
Probability Density Function and

F(x)
}
F(b)
P(a  X  b)=F(b) - F(a)

F(a)
0
a b x
f(x)
P(a  X  b) = Area
under
f(x) between a and b
= F(b) - F(a)
x
0 a b
Standard Distributions
Six continuous distributions :
1. uniform distribution
2. normal distribution
3. exponential distribution
4. t distribution
5. chi-square distribution
6. F distribution
3-71
Uniform Distribution
x1 x2
3-72
Problem
Suppose a production line is set up to manufacture machine braces
in lots of five per minute during a shift. When the lots are
weighed, variation among the weights is detected, with lot weights
ranging from 41 to 47 grams in a uniform distribution.
a) What is the mean and standard deviation ?
b) Determine the probability that a lot weighs between 42 and 45
grams.
3-73
Problem
Problem
Suppose the amount of time it takes to assemble a plastic
module ranges from 27 to 39 seconds and that assembly
times are uniformly distributed. What is the probability that
a given assembly will take between 30 and 35 seconds?
Fewer than 30 seconds?
Problem
Suppose the amount of time it takes to assemble a plastic
module ranges from 27 to 39 seconds and that assembly
times are uniformly distributed. What is the probability that
a given assembly will take between 30 and 35 seconds?
Fewer than 30 seconds?
Problem
According to the National Association of Insurance
Commissioners, the average annual cost for automobile
insurance in the United States in a recent year was $691.
Suppose automobile insurance costs are uniformly
distributed in the United States with a range of from $200 to
$1,182. What is the standard deviation of this uniform
distribution? What is the height of the distribution? What
is the probability that a person’s annual cost for
automobile insurance in the United States is between
$410 and $825?
Problem
Exponential Distribution
• It is closely related to the Poisson distribution. Whereas the Poisson
distribution is discrete and describes random occurrences over some
interval, the exponential distribution is continuous and describes a
probability distribution of the times between random occurrences
• Characteristics of the exponential distribution.
■ It is a continuous distribution.
■ It is a family of distributions.
■ It is skewed to the right.
■ The x values range from zero to infinity.
■ Its apex is always at x = 0.
■ The curve steadily decreases as x gets larger
Exponential Distribution
Mean = 1 / λ
σ=1/λ
x0 is the fraction of the interval or the number of intervals between arrivals

Problem
Arrivals at a bank are Poisson distributed with a of 1.2 customers every
minute. What is the average time between arrivals and what is the
probability that at least 2 minutes will elapse between one arrival and
the next arrival?
Problem
Arrivals at a bank are Poisson distributed with a of 1.2 customers every
minute. What is the average time between arrivals and what is the
probability that at least 2 minutes will elapse between one arrival and
the next arrival?
Sol.
Interarrival times of random arrivals are exponentially distributed
The mean of this exponential distribution is μ = 1 /λ = 1 / 1.2 = 0.833 minute

= 50 seconds
Problem
A manufacturing firm has been involved in statistical quality
control for several years. As part of the production process,
parts are randomly selected and tested. From the records of
these tests, it has been established that a defective part
occurs in a pattern that is Poisson distributed on the average
of 1.38 defects every 20 minutes during production runs.
Use this information to determine the probability that less
than 15 minutes will elapse between any two defects.
Problem
A manufacturing firm has been involved in statistical quality
control for several years. As part of the production process,
parts are randomly selected and tested. From the records of
these tests, it has been established that a defective part
occurs in a pattern that is Poisson distributed on the average
of 1.38 defects every 20 minutes during production runs.
Use this information to determine the probability that less
than 15 minutes will elapse between any two defects.
Sol.
It is .7246 of the interval, or (.7246)(20 minutes) = 14.49 minutes, between

defects.
x0 = 15 / 20, or .75 of an interval.
P(x0 < x0) = 1 – 0.3552 = 0.6448

4-1
The Normal Distribution

4-2
Introduction
As n increases, the binomial distribution approaches a ...
n=6 n = 10 n = 14
Binomial Distribution: n=6, p=.5 Binomial Distribution: n=10, p=.5 Binomial Distribution: n=14, p=.5
0.3 0.3 0.3
0.2 0.2 0.2

x
(
)
P(x
P(x P
)
)
0.1 0.1 0.1
0.0 0.0 0.0

0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 01 11 1231 14
x x x
Normal Probability Density Function: Normal Distribution:  = 0, = 1

0.

4
x 2

 
  0.

 
f(x
3
)
f ( x) 1 e 2 2 for    x   0.
2
2 2 0.
1
-5 0 5
where e  2 .7182818 ... and   3 .14159265

x
0.
0
...
4-3
The Normal Probability Distribution

The normal distribution is described or characterized by two
parameters: the mean, μ , and the standard deviation,σ .
The normal probability density function: Normal Distribution:  = 0, = 1
0.
 2
4
x


 
 0.
2 2
3
for
f(x
f ( x ) 1 e x
)
0.
2 2
2
0.
where e  2.7182818 ... and   3.14159265 ... 1 -5 0

x
5
0.0
4-4
Properties of the Normal Distribution
 The normal distribution exhibits the following

characteristics.
■ It is a continuous distribution.
■ It is a symmetrical distribution about its
mean.
■ It is asymptotic to the horizontal axis.
■ It is unimodal.
■ It is a family of curves.
■ Area under the curve is 1.
4-5
The Standard Normal Distribution
The standard normal random variable, Z, is the normal random

variable with mean  = 0 and standard deviation  = 1:
Z~N(0,12).
Standard Normal Distribution
0.4
0.3
=1
f(z)
{
0.2
0.1
0.0
-5 -4 -3 -2 -1 0 1 2 3 4 5
=0
Z
4-6
Finding Probabilities of the

Standard Normal Distribution: P(0
< Z < 1.56) Standard Normal Probabilities
Standard Normal z
0.0
.00
0.0000
.01
0.0040
.02
0.0080
.03
0.0120
.04
0.0160
.05
0.0199
.06
0.0239
.07
0.0279
.08
0.0319
.09
0.0359
0.
Distribution 0.1 0.0398 0.0438 0.0478 0.0517 0.0557 0.0596 0.0636 0.0675 0.0714 0.0753
0.2 0.0793 0.0832 0.0871 0.0910 0.0948 0.0987 0.1026 0.1064 0.1103 0.1141
4
0.3 0.1179 0.1217 0.1255 0.1293 0.1331 0.1368 0.1406 0.1443 0.1480 0.1517
0.4 0.1554 0.1591 0.1628 0.1664 0.1700 0.1736 0.1772 0.1808 0.1844 0.1879
0.5 0.1915 0.1950 0.1985 0.2019 0.2054 0.2088 0.2123 0.2157 0.2190 0.2224
0.
0.6 0.2257 0.2291 0.2324 0.2357 0.2389 0.2422 0.2454 0.2486 0.2517 0.2549
3
0.7 0.2580 0.2611 0.2642 0.2673 0.2704 0.2734 0.2764 0.2794 0.2823 0.2852
f(z
0.8 0.2881 0.2910 0.2939 0.2967 0.2995 0.3023 0.3051 0.3078 0.3106 0.3133
0. 0.9 0.3159 0.3186 0.3212 0.3238 0.3264 0.3289 0.3315 0.3340 0.3365 0.3389
1.0 0.3413 0.3438 0.3461 0.3485 0.3508 0.3531 0.3554 0.3577 0.3599 0.3621
2 1.56 1.1 0.3643 0.3665 0.3686 0.3708 0.3729 0.3749 0.3770 0.3790 0.3810 0.3830
{
1.2 0.3849 0.3869 0.3888 0.3907 0.3925 0.3944 0.3962 0.3980 0.3997 0.4015
0.0
0. 1.3 0.4032 0.4049 0.4066 0.4082 0.4099 0.4115 0.4131 0.4147 0.4162 0.4177
1 -5 -4 -3 -2 -1 0 1 2 3 1.4 0.4192 0.4207 0.4222 0.4236 0.4251 0.4265 0.4279 0.4292 0.4306 0.4319
4 5 1.5 0.4332 0.4345 0.4357 0.4370 0.4382 0.4394 0.4406 0.4418 0.4429 0.4441
1.6 0.4452 0.4463 0.4474 0.4484 0.4495 0.4505 0.4515 0.4525 0.4535 0.4545
Z
1.7 0.4554 0.4564 0.4573 0.4582 0.4591 0.4599 0.4608 0.4616 0.4625 0.4633
1.8 0.4641 0.4649 0.4656 0.4664 0.4671 0.4678 0.4686 0.4693 0.4699 0.4706
1.9 0.4713 0.4719 0.4726 0.4732 0.4738 0.4744 0.4750 0.4756 0.4761 0.4767
2.0 0.4772 0.4778 0.4783 0.4788 0.4793 0.4798 0.4803 0.4808 0.4812 0.4817
P(0  z  1.56) = 0.4406

2.1 0.4821 0.4826 0.4830 0.4834 0.4838 0.4842 0.4846 0.4850 0.4854 0.4857
2.2 0.4861 0.4864 0.4868 0.4871 0.4875 0.4878 0.4881 0.4884 0.4887 0.4890
2.3 0.4893 0.4896 0.4898 0.4901 0.4904 0.4906 0.4909 0.4911 0.4913 0.4916
2.4 0.4918 0.4920 0.4922 0.4925 0.4927 0.4929 0.4931 0.4932 0.4934 0.4936
2.5 0.4938 0.4940 0.4941 0.4943 0.4945 0.4946 0.4948 0.4949 0.4951 0.4952
2.6 0.4953 0.4955 0.4956 0.4957 0.4959 0.4960 0.4961 0.4962 0.4963 0.4964
2.7 0.4965 0.4966 0.4967 0.4968 0.4969 0.4970 0.4971 0.4972 0.4973 0.4974
2.8 0.4974 0.4975 0.4976 0.4977 0.4977 0.4978 0.4979 0.4979 0.4980 0.4981
2.9 0.4981 0.4982 0.4982 0.4983 0.4984 0.4984 0.4985 0.4985 0.4986 0.4986
3.0 0.4987 0.4987 0.4987 0.4988 0.4988 0.4989 0.4989 0.4989 0.4990 0.4990
4-7

Standard Normal Distribution: P(Z
< -2.47) z ... .06 .08
To find P(Z<-2.47): . .
P(Z < -2.47) = .5 - P(0 < Z < . .07 .
. .
2.47)
= .5 - .4932 = 0.0068 .
2.5 ... 0.4948
. 0.4949
0.4951
. .
. . .
.2.3 ... 0.4909 0.4911
Area to the left of -2.47 0.4913
0. 2.4 ... 0.4931 0.4932 0.4934
P(Z < -2.47) = .5 - 0.4932 4
= 0.0068 0. Table area for 2.47

3
P(0 < Z < 2.47) = 0.4932
f(z
0.
)
0.1
0.0
-5 -4 -3 -2 -1 0 1 2 3
4 5
Z
4-8

Standard Normal Distribution: P(1<
Z < 2)
4-9

Standard Normal Distribution: P(1<
Z < 2)
To find P(1  Z  2): z
.
.00
.
...
. .
1. Find table area for 2.00 .
0.9
.
0.3159 ...
1.0 0.3413 ...
F(2) = P(Z  2.00) = .5 + .4772 =.9772 1.1
.
0.3643
.
...
2. Find table area for 1.00 .

.
.
.
1.9 0.4713 ...
F(1) = P(Z  1.00) = .5 + .3413 = .8413 2.0
2.1
0.4772
0.4821
...
...
. .
3. P(1  Z  2.00) = P(Z  2.00) - P(Z  1.00) .
.
.
.
= .9772 - .8413 = 0.1359

0.4
0.3
Area between 1 and 2
P(1  Z  2) = .9772 - .8413 = 0.1359
f(z
0.2
)
0.1
0.0
-5 -4 -3 -2 -1 0 1 2 3
4 5
Z
4-10
Finding Values of the Standard Normal
Random Variable: P(0 < Z < 1.28) and
P (Z< 1.28)
4-11
Finding Values of the Standard Normal
Random Variable: P(0 < Z < 1.28) and
P (Z< 1.28)
z .00 .01 .02 .03 .04 .05 .06 .07 .08 .09
To find z such that 0.0
0.1
0.0000
0.0398
0.0040
0.0438
0.0080
0.0478
0.0120
0.0517
0.0160
0.0557
0.0199
0.0596
0.0239
0.0636
0.0279
0.0675
0.0319
0.0714
0.0359
0.0753
0.2 0.0793 0.0832 0.0871 0.0910 0.0948 0.0987 0.1026 0.1064 0.1103 0.1141
0.3 0.1179 0.1217 0.1255 0.1293 0.1331 0.1368 0.1406 0.1443 0.1480 0.1517
P(0  Z  z) = .40:
0.4 0.1554 0.1591 0.1628 0.1664 0.1700 0.1736 0.1772 0.1808 0.1844 0.1879
0.5 0.1915 0.1950 0.1985 0.2019 0.2054 0.2088 0.2123 0.2157 0.2190 0.2224
0.6 0.2257 0.2291 0.2324 0.2357 0.2389 0.2422 0.2454 0.2486 0.2517 0.2549
0.7 0.2580 0.2611 0.2642 0.2673 0.2704 0.2734 0.2764 0.2794 0.2823 0.2852
0.8 0.2881 0.2910 0.2939 0.2967 0.2995 0.3023 0.3051 0.3078 0.3106 0.3133
1. Find a probability as close as

0.9 0.3159 0.3186 0.3212 0.3238 0.3264 0.3289 0.3315 0.3340 0.3365 0.3389
1.0 0.3413 0.3438 0.3461 0.3485 0.3508 0.3531 0.3554 0.3577 0.3599 0.3621
1.1 0.3643 0.3665 0.3686 0.3708 0.3729 0.3749 0.3770 0.3790 0.3810 0.3830
possible to .40 in the table of 1.2
1.3
0.3849
0.4032
0.3869
0.4049
0.3888
0.4066
0.3907
0.4082
0.3925
0.4099
0.3944
0.4115
0.3962
0.4131
0.3980
0.4147
0.3997
0.4162
0.4015
0.4177
standard normal probabilities. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . . . . . . . . . .
2. Then determine the value of z Standard Normal

from the corresponding row Distribution
0.
and column. Area to the left of 0 = .50 4
Area = .40 (.3997)
P(z  0) = .50 0.
3
P(0  Z  1.28)  .40
f(z
0.
)
Also, since P(Z  0) = .

0.
1
50 0.
0 -5 -4 -3 -2 -1 0 1 2 3 4 5
P(Z  1.28)  .90 Z Z = 1.28
4-12
99% Interval around the Mean
To have .99 in the center of the distribution, there

z .04 .05 .06 .07 .08 .09
should be (1/2)(1-.99) = (1/2)(.01) = .005 in each . . . . . . .
. . . . . . .
tail of the distribution, and (1/2)(.99) = .495 in .
2.4 ...
.
0.4927
.
0.4929
.
0.4931
.
0.4932
.
0.4934
.
0.4936
each half of the .99 interval. That is: 2.5 ...
2.6 ...
0.4945
0.4959
0.4946
0.4960
0.4948
0.4961
0.4949
0.4962
0.4951
0.4963
0.4952
0.4964
. . . . . . .
. . . . . . .
P(0  Z  z.005 ) = . . . . . . . .
495
Look to the table of standard normal probabilities Total area in center = .99
to find that: Area in center left = .495
0.4
  z.005   Area in center right = .495
z.005  
0.3
f(z
0.2
)
P(-.2575  Z  Area in left tail = .005
0.1
Area in right tail = .005
) = .99 0.0

-5 -4 -3 -2 -1 0 1 2 3 4 5
Z
-z.005 z.005
-2.575 2.575
4-13
Normal Probability Distributions

All of these are normal probability density functions, though each has a different
mean and variance but the area under the curve for –kσ < x < kσ is same
Normal Distribution:  =40, =1 Normal Distribution:  =30, =5 Normal Distribution:  =50, =3
0.4 0.2
0.2
0.3
f(w
f(x
f(y
0.2 0.1 0.1
)
)
0.1
0.0 0.0 0.0

35 40 45 0 10 20 30 40 50 60 35 45 50 55 65
w x y
W~N(40,1) X~N(30,25) Y~N(50,9)

Normal Distribution:  =0, =1
0.4 Consider:
0.3
The probability in each
P(39  W  41) case is an area under a
f(z
0.2
)
0.1
P(25  X  35) normal probability density
0.0
P(47  Y  53) function.
P(-1  Z  1)
-5 0 5
z
Z~N(0,1)
4-14
The Transformation of Normal

Random Variables
The area within k of the mean is the same for all normal random variables. So an area
under any normal distribution is equivalent to an area under the standard normal. In this
example: P(40  X    P(-1  Z     since    and   
The transformation of X to Z:
X   x Normal Distribution:  =50,
Z  =10

x 0.0
7
Transformation 0.0
6
f(x
0.0
(1) Subtraction: (X - x)
)
5
0.0
3
0.0
0.0 =10
{
Standard Normal Distribution 4
2
0.0
0.4 1
0 10 20 30 40 50 60 70 80 90 100
0.0
0 X
0.3
f(z
0.2
)
(2) Division by x)

The inverse transformation of Z to X:
{
0.1 1.0
XxZx
0.0
-5 -4 -3 -2 -1 0 1 2 3 4 5
Z
4-15
The Transformation of Normal

Random Variables
The transformation of X to Z: The inverse transformation of Z to X:

X x
Z  X    Z
x x
x
The transformation of X to Z, where a and b are numbers::
 a 

P( X  a)  P Z   
 b
P( X  b)  P  Z   
  
   X  b)  P a
P(a Zb 




4-16
Using the Normal Transformation

X~N(160,302). Find P (100 < X < 180)
4-17

X~N(160,302). Find P (100 < X < 180)
P (100  X  180 )
 100
  180   
 P    
X  

 100  160 180  160 
 P  30 Z  30 
 P  2  Z  .6666
 0.4772  0.2475  0.7247

4-18

X~N(127,222). Find P (X < 150)
4-19

X~N(127,222). Find P (X < 150)
P ( X  150 )
 X  150   
 P
  

150  127 
 P Z  
22
 P  Z  1.045
 0.5  0.3520 
0.8520
4-20

X~N(383,122). Find P (394 < X < 399)
4-21

X~N(383,122). Find P (394 < X < 399)
P(394  X  399)
 394    X    399 
 P    
 394  383 399 383
 P 12 Z 
12
 P 0.9166 Z  1.333
 0.4088 0.3203 0.0885

4-22
Normal Probabilities (Empirical Rule)
• The probability that a normal random S tand ard N o rm al D is trib utio

n
variable will be within 1 standard 0.
4
deviation from its mean (on either 0.
side) is 0.6826, or approximately 0.68. 3
f(z)
0.
• The probability that a normal

2
0.
random variable will be within 2 1
standard deviations from its mean is 0.

0 -5 -4 -3 -2 -1 0 1 2 3 4 5
0.9544, or approximately 0.95. Z
• The probability that a normal

random variable will be within 3
standard deviation from its mean is
0.9974.
Six Sigma
 Is the relentless and rigorous pursuit of the reduction
of variation in all critical processes to achieve
continuous and breakthrough improvements that
impact the bottom line of the organization and
increase customer satisfaction.
 It is an organizational initiative designed to create
manufacturing, service and administrative processes
that produce approximately 3.4 defects per million.
Opportunities (DPMO).
Process Capability
Process Capability
4-26
The Inverse Transformation
X~N(124,122)
P(X > x) = 0.10 .
Find x.
4-27
The Inverse Transformation
X~N(124,122)
P(X > x) = 0.10 .
Find x.
X~N(124,122)
P(X > x) = 0.10 and P(Z > 1.28)  0.10
x =  + z = 124 + (1.28)(12) = 139.36
4-28
The Inverse Transformation (Continued)
X~N(5.7,0.52)
P(X >
x)=0.01
Find x.
4-29
X~N(5.7,0.52)
P(X >
x)=0.01
Find x.
X~N(5.7,0.52)
P(X > x)=0.01 and P(Z > 2.33) 
0.01 x =  + z = 5.7 + (2.33)(0.5) =
6.865
4-30

X~N(2450,4002)
P(a<X<b)=0.95
Find a, b.
4-31

X~N(2450,4002)
P(a<X<b)=0.95
Find a, b.
and P(-1.96<Z<1.96) 0.95
x =   z = 2450 ± (1.96)(400) = 2450 ±784=(1666,3234)
P(1666 < X < 3234) = 0.95
4-32
Approximating a Binomial
Probability Using the Normal
Distribution
For large n values, the binomial distribution is cumbersome to
analyze without a computer. Binomial Probability Tables goes only
to n = 25. The normal distribution is a good approximation for
binomial distribution problems for large values of n.
Test for validation:

1. Does the interval μ + 3σ lie between 0 and n?
For a normal curve approximation of a binomial distribution
problem to be acceptable, all possible x values should be
between 0 and n, which are the lower and upper limits,
respectively, of a binomial distribution.
2. Approximation is good enough if both n . p > 5 and n . q > 5
3-33
Shape of the Binomial Distribution

p = 0.1 p = 0.3 p = 0.5
0.7 0.7 0.7
0.6 0.6 0.6
0.5 0.5 0.5
n=4 0.4 0.4 0.4

P(x
P(x
P(x
)
)
0.3 0.3 0.3
0.2 0.2 0.2
0.1 0.1 0.1
0.0 0.0 0.0

0 1 2 3 4 0 1 2 3 4 0 1 2 3 4
x x x
Binomial Probability: n=10 p=0.1 Binomial Probability: n=10 p=0.3 Binomial Probabil i ty: n=10 p=0.5
0. 5 0. 5 0. 5
0. 4 0. 4 0. 4
n = 10 0. 3 0. 3 0. 3
P(x)
P(x
P(x
)
)
0. 2 0. 2 0. 2
0. 1 0. 1 0. 1
0. 0 0. 0 0. 0
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
x
x x
0.2 0.2 0.2
n = 20
P(x
P(x
P(x
)
)
0.1 0.1 0.1
0.0 0.0 0.0

0 1 2 3 4 5 6 7 8 9 1011121314151617181920 0 1 2 3 4 5 6 7 8 9 1011121314151617181920 0 1 2 3 4 5 6 7 8 9 1011121314151617181920
x x x
Binomial distributions become more symmetric as n increases and as p 0.5.

4-34
The Normal Approximation of

Binomial Distribution
The normal distribution with  = 5.5 and  = 1.6583 is a
closer approximation to the binomial with n = 11 and p = 0.50.
Normal Distribution:  = 5.5, = 1.6583 Binomial Distribution: n = 11, p = 0.50
0.3
0.2
0.2
P(x
f(x
)
)
0.1
0.1
0.0 0.0
0 5 10 0 1 2 3 4 5 6 7 8 9 10 11
X X
P(x < 4.5) = P(x  4) = 0.2744

0.2732
Problem
Binomial distribution problem
Problem
Binomial distribution problem
The binomial problem becomes a normal curve problem.
Test is made to determine whether the normal curve

sufficiently fits this binomial distribution to justify the use of
the normal curve.
This interval is between 0 and 60, so the approximation is

sufficient to allow use of the normal curve
Correcting for Continuity
• A correction of +.50 or -.50 or +.50, depending on the problem,
is required. This correction ensures that most of the binomial
problem’s information is correctly transferred to the normal
curve analysis. This correction is called the correction for
continuity, which is made during conversion of a discrete
distribution into a continuous distribution.
Conversion to a continuous normal curve problem yields
= 0.0336
Problem
Problem
A binomial probability at x = 12 translates to a continuous normal

curve area that lies between 11.5 and 12.5.
P ( 11.5 < x < 12.5 | μ =10 and σ = 2.45)

Problem
P ( 11.5 < x < 12.5 | μ =10 and σ = 2.45) = .1170
Problem
Problem
P ( x < 26.5 | μ = 37 and σ = 4.83) = .0150
5-1
Sampling and Sampling Distributions

5-2
Using Statistics
• Statistical Inference:
 Predict and forecast values of On basis of sample
population parameters... statistics derived from
 Test hypotheses about values limited and incomplete
of population sample information.
parameters...Make decisions..
Make
generalizations
about the On the basis of
characteristics of observations of a
a population... sample, a part
of a population
Sampling - Frame
• Every research study has a target population that consists of
the individuals, institutions, or entities that are the object of
investigation. The sample is taken from a population list, map,
• directory, or other source used to represent the population. This
list, map, or directory is called the frame.
• School lists, trade association lists, or even lists sold by list
brokers.
• Frames that have overregistration contain the target population
units plus some additional units.
• Frames that have underregistration contain fewer units than does
the target population.
• Sampling is done from the frame, not the target population.
• In theory, the target population and the frame are the same.
• In reality, a business researcher’s goal is to minimize the
differences between the frame and the target
Types of Sampling
The two main types of sampling are random and nonrandom.
• In random sampling every unit of the population has the same
probability of being selected into the sample. Random sampling
implies that chance enters into the process of selection.
• winners of nationwide magazine sweepstakes or numbers
selected as state lottery winners are selected by some random
draw of numbers
• In nonrandom sampling not every unit of the population has the
same probability of being selected into the sample. Members of
nonrandom samples are not selected by chance.
• People might be selected because they are at the right place at the
right time or because they know the people conducting the
research.
Random Sampling Techniques
The four basic random sampling

techniques are
1.simple random sampling
2.stratified random sampling
3.systematic random sampling, and
4.cluster (or area) random sampling
Non-random Sampling
Sampling techniques used to select elements from the
population by any mechanism that does not involve a
random selection process. Four nonrandom sampling
techniques are,
1. Convenience Sampling
2. Judgement Sampling
3. Quota Sampling
4. Snowball sampling
Sampling Error
Sampling error occurs when the sample is not representative of the
population. When random sampling techniques are used to select
elements for the sample, sampling error occurs by chance. Many times
the statistic computed on the sample is not an accurate estimate of
the population parameter because the sample was not representative of
the population. This result is caused by sampling error. With random
samples, sampling error can be computed and analyzed.
Non-sampling Errors
All errors other than sampling errors are nonsampling errors. The
many possible nonsampling errors include missing data, recording
errors, input processing errors, and analysis errors. Other nonsampling
errors result from the measurement instrument, such as errors
of unclear definitions, defective questionnaires, response errors and
poorly conceived concepts. Improper definition of the frame is a
nonsampling error. In many cases, finding a frame that perfectly fits the
population is impossible. Virtually no statistical method is available to
measure or control for nonsampling errors. The researcher must
eliminate these errors through carefully planning and executing the
research study.
5-9
A Population Distribution, a Sample from a

Population, and the Population and Sample
Means
Population mean ()
Frequency distribution of
the population
X X X X X X X
X X X X X X
X X X X X
Sample points
Sample mean ( X)
5-10
Sampling Distributions
• The sampling distribution of a statistic is the probability

distribution of all possible values the statistic may assume,
when computed from random samples of the same size, drawn
from a specified population.
• The sampling distribution of X is the probability distribution of
all possible values the random variable X may assume when a
sample of size n is taken from a specified population.
5-11
Sampling Distributions (Continued)
Uniform population of integers from 1 to 8:

X P(X) XP(X) (X-x) (X-x)2 P(X)(X-x)2 Uniform Distribution (1,8)
1 0.125 0.125 -3.5 12.25 1.53125 0.2
2 0.125 0.250 -2.5 6.25 0.78125

3 0.125 0.375 -1.5 2.25 0.28125
4 0.125 0.500 -0.5 0.25 0.03125
P(X)
5 0.125 0.625 0.5 0.25 0.03125 0.1
6 0.125 0.750 1.5 2.25 0.28125
7 0.125 0.875 2.5 6.25 0.78125
8 0.125 1.000 3.5 12.25 1.53125
0.0
1.000 4.500 5.25000 1 2 3 4 5 6 7 8
X
E(X) =  = 4.5
V(X) = 2 = 5.25
SD(X) =  = 2.2913
5-12

◼ There are 8*8 = 64 different but Each of these samples has a
equally-likely samples of size 2 sample mean. For example, the
that can be drawn (with mean of the sample (1,4) is
replacement) from a uniform 2.5, and the mean of the
population of the integers sample (8,4) is 6.
from
1 to 8:
Samples of Size 2 from Uniform (1,8) Sample Means from Uniform (1,8), n
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
1 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
2 2,1 2,2 2,3 2,4 2,5 2,6 2,7 2,8 2 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
3 3,1 3,2 3,3 3,4 3,5 3,6 3,7 3,8 3 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
4 4,1 4,2 4,3 4,4 4,5 4,6 4,7 4,8 4 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
5 5,1 5,2 5,3 5,4 5,5 5,6 5,7 5,8 5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
6 6,1 6,2 6,3 6,4 6,5 6,6 6,7 6,8 6 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
7 7,1 7,2 7,3 7,4 7,5 7,6 7,7 7,8 7 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
8 8,1 8,2 8,3 8,4 8,5 8,6 8,7 8,8 8 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
5-13

The probability distribution of the sample mean is called
the
sampling distribution
Sampling Distribution of Mean
of the the sample mean.
X P(X) XP(X) X-X (X-X)2 P(X)(X-X)2
Sampling Distribution of the Mean
1.0 0.015625 0.015625 -3.5 12.25 0.191406
1.5 0.031250 0.046875 -3.0 9.00 0.281250
0.10
2.0 0.046875 0.093750 -2.5 6.25 0.292969
2.5 0.062500 0.156250 -2.0 4.00 0.250000
P(X)
3.0 0.078125 0.234375 -1.5 2.25 0.175781 0.05
3.5 0.093750 0.328125 -1.0 1.00 0.093750
4.0 0.109375 0.437500 -0.5 0.25 0.027344
4.5 0.125000 0.562500 0.0 0.00 0.000000 0.00
5.0 0.109375 0.546875 0.5 0.25 0.027344 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
8.0
5.5 0.093750 0.515625 1.0 1.00 0.093750 X
6.0 0.078125 0.468750 1.5 2.25 0.175781
6.5 0.062500 0.406250 2.0 4.00 0.250000 E ( X̅ )  μx̅  4.5
7.0
7.5
0.046875
0.031250
0.328125
0.234375
2.5
3.0
6.25
9.00
0.292969
0.281250
V ( X̅ )  σ2x̅  2.625
8.0 0.015625 0.125000 3.5 12.25 0.191406 SD( X̅ )  σx̅ 
1.000000 4.500000 2.625000 1.6202
5-14
Properties of the Sampling

Distribution of the Sample Mean
• Comparing the population Uniform POPULATION Distribution (1,8)

0.2
distribution and the sampling

distribution of the mean:
P(X)
0.1
The sampling distribution is

more bell-shaped and 0.0
symmetric. 1 2 3 4
X
5 6 7 8
 Both have the same center. Sampling Distribution of the Mean

The sampling distribution of
the mean is more compact, 0.10
with a smaller variance.

P(X
)
0.05
0.00
1.01.52.02.53.03.54.04.5 5.05.56.06.57.07.58.0
X
5-15
The Central Limit Theorem
n=5
When sampling from a population 0.2
with mean  and finite standard

5
0.2
P(
X)
0
deviation , the sampling

0.1
5
0.1 X
0
distribution of the sample mean will 0.0

5
n = 20
0.0
tend to a normal distribution 0.2
with  and standard deviation n

mean
P(
X)
0.1
as the sample size becomes large 0.0 X
(n >30). Large n
0.
4
0.
X)
f(
3
For “large enough” n: X ~ N (, / n)

2 0.
2
-

X
0.
1
0.
0
5-16
Population Distribution and the Sampling

Distribution of the Sample Mean
The expected value of the sample mean is equal to the population
mean: E( X )    
X
X
The variance of the sample mean is equal to the

population variance divided by the sample size:
 2
V(X)  2 X
X
n
The standard deviation of the sample mean, known as the
standard error of the mean, is equal to the population standard
deviation divided by the square root of the sample size:
SD( X )  
 X
n
X

5-17
Sampling from a Normal Population
When sampling from a normal population with mean  and standard

deviation , the sample mean, X, has a normal sampling distribution:
2
X ~ N (,
n
)
This means that, as the Sampling Distribution of the Sample
Mean
sample size increases, the 0.4
Sampling Distribution: n =16

sampling distribution of the 0.3
Sampling Distribution: n = 4
sample mean remains

f(X)
0.2
Sampling Distribution: n = 2
centered on the population 0.1
Normal population
Normal population
mean, but becomes more 0.0

compactly distributed around
that population mean.
The Central Limit Theorem
The central limit theorem creates the potential for applying
the normal distribution to many problems when sample size is
sufficiently large. Sample means that have been computed
for random samples drawn from normally distributed
populations are
normally distributed.
However, the real advantage of the central limit theorem
comes when
sample data drawn from populations not normally distributed or
from populations of unknown shape also can be analyzed by using
the normal distribution because the sample means are normally
5-19
The Central Limit Theorem Applies to

Sampling Distributions from Any Population
Normal Uniform Skewed General
Population
n=2
n = 30
 X  X  X  X
Problem
The mean expenditure per customer at a tire store is $85.00, with a
standard deviation of $9.00. If a random sample of 40 customers is
taken, what is the probability that the sample average expenditure
per customer for this sample will be $87.00 or more?
Solution
7.93% of the time, a random sample of 40 customers from this

population will yield a sample mean expenditure of $87.00 or
more.
Problem
Suppose that during any hour in a large department store, the average
number of shoppers is 448, with a standard deviation of 21 shoppers.
What is the probability that a random sample of 49 different shopping
hours will yield a sample mean between 441 and 446 shoppers?
Solution
The probability of a value being between z = -2.33 and -0.67 is

.2415; that is, there is a 24.15% chance of randomly selecting
49 hourly periods for which the sample mean is between 441
and 446 shoppers.
5-24
Problem
Mercury makes a 2.4 liter V-6 engine, the Laser XRi, used in
speedboats. The company’s engineers believe the engine
delivers an average power of 220 horsepower and that the
standard deviation of power delivered is 15 HP. A potential
buyer intends to sample 100 engines (each engine is to be run a
single time). What is the probability that the sample mean will
be less than 217HP?
5-25
Solution
 
P( X  217)  P  X    217   
  
 n  
n
   
 217   217 
 P Z  
  P Z 
 2201 5  22015
 
 100 10 
 
 P(Z  2) 
0.0228
Sampling from a Finite Population
In cases of a finite population, a statistical adjustment can be
made to the z formula for sample means. The adjustment is
called the finite correction factor
Problem
A production company’s 350 hourly employees average 37.6 years
of age, with a standard deviation of 8.3 years. If a random sample
of 45 hourly employees is taken, what is the probability that the
sample will have an average age of less than 40 years?
Solution
The probability of getting a sample average age of less than 40

years is .4808 + .5000 = .9808. Had the finite correction factor
not been used, the z value would have been 1.94, and the final
answer would have been .9738.
Proportion
If research produces measurable data such

as weight, distance, time, and income, the
sample mean is often the statistic of
choice.
However,
If research results in countable items such
as how many then proportion is often the
statistic of choice.
5-30
Population and Sample Proportions
• The population proportion is equal to the number of elements

in the population belonging to the category of interest, divided
by the total number of elements in the population:
X
p N
• The sample proportion is the number of elements in the

sample belonging to the category of interest, divided by
the sample size:
p$  n
x
5-31
The Sampling Distribution of the

Sample Proportion, p$
n=2, p = 0.3
The sample proportion is the percentage 0 .5
0 .4
of successes in n binomial trials. It is the X

(
0 .3
)
P
0 .2
number of successes, X, divided by the 0 .1
number of trials, n.
0 .0
01 2
n=10,p=0.3
X
0.3
$ p
Sample proportion:
0.2
P(X
N
)
0.1
0.0
As the sample size, n, increases, the sampling

0 1 2 3 4 5 6 7 8 9 10
X
distribution of p$ approaches a normal n=15, p = 0.3
distribution with mean p and standard 0.2
deviation X
(
)
P 0.1
p(1 
p) n 0.0
0 1 2 3 4 5 6 7 8 9 1011 21 31 41 51
0 1 2 3 4 5 6 7 8 9 10 11 12 13 1415
X
15 1515 15 15 15 15 15 151515 1515 15 1515 ^
when n.p > 5 and n.q > p

Problem
Suppose 60% of the electrical contractors in a region use a
particular brand of wire. What is the probability of taking a random
sample of size 120 from these electrical contractors and finding that
.50 or less use that brand of wire ?
Solution
A researcher would have probability of .0125 in finding that 50%

or less of a sample of 120 contractors use a given brand of wire.
Problem
If 10% of a population of
parts is defective, what is the
probability of randomly
selecting 80 parts and finding
that 12 or more parts are
defective?
Solution
Thus, about 6.81% of the time, 12 or more defective parts

would appear in a random sample of 80 parts when the
population proportion is .10.
Problem
In recent years, convertible sports coupes have become very
popular in Japan. Toyota is currently shipping Celicas to Los
Angeles, where a customizer does a roof lift and ships them back
to Japan. Suppose that 25% of all Japanese in a given income
and lifestyle category are interested in buying Celica
convertibles. A random sample of 100 Japanese consumers
in the category of interest is to be selected.
What is the probability that at least 20% of those in the sample
will express an interest in a Celica convertible?
5-37
Solution
n  100  
p  0.25 P( p$  0.20)  P p$  p  .20  p 
 p(1  p) p(1 
np  (100)(0.25)  25  E ( p$ )  p) n n

 
p(1  p) 
(.25)(.75)
 .20 .25
  .05
n 100
 0.001875  V ( p$ ) 
P z

(.25)(.75)   
 P z
 .0433 

p(1  p)   100
0.001875  0.04330127  SD( p$ ) 
n 
 P ( z  1.15)  0.8749
5-38
Sample Statistics as Estimators of

Population Parameters
◼ A sample statistic is a
numerical measure of a
summary
characteristic of a
sample.
A population parameter
is a numerical measure of
a summary characteristic
of a population.
5-39
Sample Statistics as Estimators of

Population Parameters
• An estimator of a population parameter is a sample
statistic used to estimate or predict the population
parameter.
• An estimate of a parameter is a particular numerical value
of a sample statistic obtained through sampling.
• A point estimate is a statistic taken from a sample that is
used to estimate a population parameter.
• A interval estimate (confidence interval) is a range of
values within which the analyst can declare, with
5-40
Estimators
• The sample mean, X , is the most common

estimator of the population mean, 
• The sample variance, s2, is the most
common estimator of the population
variance, 2.
• The sample standard deviation, s, is the most
common estimator of the population standard
deviation, .
• The sample proportion, pˆ, is the most
common estimator of the population
5-41
Estimators
5-42
Unbiasedness
An estimator is said to be unbiased if its expected value is equal to

the population parameter it estimates.
For example, E(X)= so the sample mean is an unbiased estimator of

the population mean. Unbiasedness is an average or long-run
property. The mean of any single sample will probably not equal the
population mean, but the average of the means of repeated
independent samples from a population will equal the population
mean.
Any systematic deviation of the estimator from the

population parameter of interest is called a bias.
5-43
Unbiased and Biased Estimators
{
Bias
An unbiased estimator is on A biased estimator is

target on average. off target on
average.
5-44
Efficiency
An estimator is efficient if it has a relatively small variance

(and standard deviation).
An efficient estimator is, An inefficient estimator is, on

on average, closer to the average, farther from the
parameter being estimated.. parameter being estimated.
5-45
Consistency and Sufficiency
An estimator is said to be consistent if its probability of being

close to the parameter it estimates increases as the sample size
increases.
Consistency
n = 10 n = 100
An estimator is said to be sufficient if it contains all the

information in the data about the parameter it estimates.
5-46
Properties of the Sample Mean
For a normal population, both the sample mean and

sample median are unbiased estimators of the
population mean, but the sample mean is both more
efficient (because it has a smaller variance), and
sufficient. Every observation in the sample is used in
the calculation of the sample mean, but only the middle
value is used to find the sample median.
In general, the sample mean is the best estimator of the
population mean. The sample mean is the most
efficient unbiased estimator of the population mean. It
is also a consistent estimator.
5-47
Properties of the Sample Variance
The sample variance (the sum of the squared deviations from the
sample mean divided by (n-1) is an unbiased estimator of the
population variance. In contrast, the average squared deviation
from the sample mean is a biased (though consistent) estimator of
the population variance.
E (s2 )    ( x  x)   
E  2
 (n  1)
  ( x  x) 2 
E  2
 n 
6-1
Confidence Intervals
6-2
Using Statistics
• Consider the following statements:

x = 550
• A single-valued estimate that conveys little information
about the actual value of the population mean.
We are 99% confident that  is in the interval
[449,551]
• An interval estimate which locates the population mean
within a narrow interval, with a high level of confidence.
We are 90% confident that  is in the interval
[400,700]
• An interval estimate which locates the population mean
within a broader interval, with a lower level of confidence.
6-3
Types of Estimators
• Point Estimate
 A single-valued estimate.
 A single element chosen from a sampling distribution.
Conveys little information about the actual value of the
population parameter, about the accuracy of the estimate.
• Confidence Interval or Interval Estimate

An interval or range of values believed to include the
unknown population parameter.
Associated with the interval is a measure of the confidence
we have that the interval does indeed contain the parameter of
interest.
6-4
Confidence Interval or Interval

Estimate
A confidence interval or interval estimate is a range or interval of

numbers believed to include an unknown population parameter.
Associated with the interval is a measure of the confidence we
have
that the interval does indeed contain the parameter of interest.
• A confidence interval or interval estimate has two components:
 A range or interval of values
 An associated level of confidence
6-5
A 95% Interval around the

Population Mean
Sampling Distribution of the
0.4 Mean
Approximately 95% of sample means
0.3
95%
can be expected to fall within the
interval 
f(x
 
0.2
  1.96 n,   1.96

.
)
n
0.1
2.5% 2.5%
Conversely,  about 2.5% can be
0.0

expected to be above   1.96 n and
x

  1.96 n   1.96 n
2.5% can be expected to be below

x
x   1.96 n .
2.5% fall below
the interval x
x
x
So 5% can be expected to fall outside
 
x 2.5% fall above  .
x
the interval the interval   1.96 n ,  n
x
1.96 
x
95% fall within

the interval
6-6
95% Intervals around the Sample

Mean
Sampling Distribution of the
0.4 Mean Approximately 95% of the intervals
x 1.96 around the sample mean can be

95%
0.3 n
expected to include the actual value of
the population mean, . (When the
f(x
0.2
)
0.1
2.5% 2.5% sample mean falls within the 95%
0.0 interval around the population mean.)
x

  1.96 n   1.96 n
x x x
x *5% of such intervals around the sample

x mean can be expected not to include the
x actual value of the population mean.
* x
x (When the sample mean falls outside the
x 95% interval around the population
x
mean.)
x
x
x
x
*
6-7
Confidence Interval for  when  is

known
⚫ If the population distribution is normal, the sampling
distribution of the mean is normal.
• If the sample is sufficiently large, regardless of the shape of the
population distribution, the sampling distribution is normal
(Central Limit Theorem).
In either case:
Standard Normal Distribution: 95% Interval
0.4
 x 
P   1.96 n     1.96   0.95 0.3
n
f(z
0.2
)
or 0.1
0.0
-4 -3 -2 -1 0 1 2 3 4
z
 x 
P x  1.96 n     1.96   0.95
n
6-8
Confidence Interval for  when

 is Known (Continued)
Before sampling, there is a 0.95probability that the interval
 1.96 
n
will include the sample mean (and 5% that it will not).
Conversely, after sampling, approximately 95% of

such intervals

x 1.96
n
will include the population mean (and 5% of them will
not).
That is, x  1.96 n is a 95% confidence interval for
.
6-9
The 95% Confidence Interval for 
A 95% confidence interval for  when  is known and sampling is

done from a normal population, or a large sample is used, is:

x  1.96 n
The quantity 1 . 9 6  n is often called the margin of error or the

sampling error.
For example, if: n = 25 A 95% confidence interval:
 = 20  20
x = 122 x  1.96  122  1.96 25
n
 122
122  (1.96)(4)
7.84
 114.16,129.84
6-10
A (1- )100% Confidence Interval for


We define z2as the z value that cuts off a right-tail area of  under
2 the standard
normal curve. (1-) is called the confidence coefficient.  is called the error
probability, and (1-)100% is called the confidence
level.
Stand ard Norm al Distrib
  
0.4
ution
P z z2  
(1 
0.3
) P   z   
 
2 

z
f(z
0.2
      
)
0.1   P  z2 z z2  (1

2
2
)
0.0
-5 -4 -3 -2 -1 0 1 2 3 4 5 (1- )100% Confidence
 Interval:
z z
2 x  z n
Z 2
2
6-11
Critical Values of z and Levels of

Confidence
(1   )
 z Stand ard Norm al Distrib ution
2 2
0.4
(1 
)
0.99 0.005 2.576 0.3
f(z
0.2
0.98 0.010 2.326
)
 
0.1
2 2
0.95 0.025 1.960 0.0
-5 -4 -3 -2 -1 0 3 4 5
0.90 0.050 1.645 1

2
z z
0.80 0.100 1.282 2
Z
2
6-12
The Level of Confidence and the

Width of the Confidence
Interval
When sampling from the same population, using a fixed sample size, the
higher the confidence level, the wider the confidence interval.
Stand ar d Nor m al Distri b uti Stan d ard Nor m al Distri b uti
on on
0.4 0.4
0.3 0.3
f(z)
f(z)
0.2 0.2
0.1 0.1
0.0 0.0
-5 -4 -3 -2 -1 0 1 2 3 4 5 -5 -4 -3 -2 -1 0 1 2 3 4 5
Z Z
80% Confidence Interval: 95% Confidence Interval:


x x
1.28 n  n
1.96
6-13
The Sample Size and the Width of

the Confidence Interval
When sampling from the same population, using a fixed confidence
level, the larger the sample size, n, the narrower the confidence
interval.
S am p ling D is trib utio n o f the M e an S am p ling D is trib utio n o f the M e
an
0 .4 0 .9
0 .8
0 .7
0 .3
0 .6
0 .5
f(x)
f(x
0 .2
)
0 .4
0 .3
0 .1 0 .2
0 .1
0 .0 0 .0
x x
95% Confidence Interval: n = 20 95% Confidence Interval: n = 40

Problem
A survey was taken of U.S. companies that do business with firms in
India. One of the questions on the survey was: Approximately how
many years has your company been trading with firms in India? A
random sample of 44 responses to this question yielded a mean of
10.455 years. Suppose the population standard deviation for this
question is 7.7 years. Using this information, construct a 90%
confidence interval for the mean number of years that a company has
been trading in India for the population of U.S. companies trading with
firms in India.
Problem
A survey was taken of U.S. companies that do business with firms in
India. One of the questions on the survey was: Approximately how
many years has your company been trading with firms in India? A
random sample of 44 responses to this question yielded a mean of
10.455 years. Suppose the population standard deviation for this
question is 7.7 years. Using this information, construct a 90%
confidence interval for the mean number of years that a company has
been trading in India for the population of U.S. companies trading with
firms in India.
Problem
Suppose a U.S. car rental firm wants to estimate the average number of
miles traveled per day by each of its cars rented in California. A random
sample of 20 cars rented in California reveals that the sample mean
travel distance per day is 85.5 miles, with a population standard
deviation of 19.3 miles. Compute a 99% confidence interval to estimate
.
Problem
Suppose a U.S. car rental firm wants to estimate the average number of
miles traveled per day by each of its cars rented in California. A random
sample of 20 cars rented in California reveals that the sample mean
travel distance per day is 85.5 miles, with a population standard
deviation of 19.3 miles. Compute a 99% confidence interval to estimate
.
Sol.
Assume that number of miles traveled per day is normally distributed
in the population. The confidence interval is
6-18
Large Sample Confidence Intervals

for the Population Mean
A large - sample (1 -  )100% confidence interval for

:
x  z
2
sn
6-19
Problem
An economist wants to estimate the average amount in checking
accounts at banks in a given region. A random sample of 100 accounts
gives x-bar = $357.60 and s = $140.00. Give a 95% confidence
interval for μ, the average amount in any checking account at a bank in
the given region.
6-20
Problem
An economist wants to estimate the average amount in checking
accounts at banks in a given region. A random sample of 100 accounts
gives x-bar = $357.60 and s = $140.00. Give a 95% confidence
interval for μ, the average amount in any checking account at a bank in
the given region.
s
x  z 0.025 140.00
n  357.60  1.96 100  357.60  27.44  330.16,385.04
6-21
Large-Sample Confidence Intervals for

the Population Proportion, p
The estimator of the population proportion, p, is the sample proportion, p. If the
sample size is large, p has an approximately normal distribution, with E( p ) = p and

n
pq
V( p ) =value,
estimated p , to
, where q= estimate
(1 - p). the
Whenstandard of p . is unknown, use the
deviationproportion
the population
For estimating p, a sample is considered large enough when both n  p an n  q are greater
than 5.
6-22
Large-Sample Confidence Intervals

for the Population Proportion, p
A large-sample (1-)100% confidence interval for the population proportion,

p:
pˆ / nˆqˆ
2
p
z
where the sample proportion, pˆ, is equal 
to the number of successes in the
sample, x, divided by the number of trials (the sample size), n, and qˆ =1-pˆ.
Problem
Coopers & Lybrand surveyed 210 chief executives of fast-
growing small companies. Only 51% of these executives
had a management succession plan in place. A
spokesperson for Cooper & Lybrand said that many
companies do not worry about management succession
unless it is an immediate problem. However, the
unexpected exit of a corporate leader can disrupt and
unfocus a company for long enough to cause it to lose its
momentum. Use the data given to compute a 92% confidence
interval to estimate the proportion of all fast-growing small
companies that have a management succession plan.
Solution
It is estimated with 92% confidence that the proportion

of the population of fast growing small companies that
have a management succession plan is between .45
and .57.
Problem
A clothing company produces men’s jeans. The jeans are
made and sold with either a regular cut or a boot cut. In an
effort to estimate the proportion of their men’s jeans
market in Oklahoma City that prefers boot-cut jeans, the
analyst takes a random sample of 212 jeans sales from the
company’s two Oklahoma City retail outlets. Only 34 of
the sales were for boot-cut jeans. Construct a 90%
confidence interval to estimate the proportion of the
population in Oklahoma City who prefer boot-cut jeans.
Solution
The analyst estimates that the population

proportion of boot-cut jeans purchases is
between .12 and .20. The level of confidence in
this result is 90%.
6-27
Example
A marketing research firm wants to estimate the share that foreign

companies have in the American market for certain products. A
random sample of 100 consumers is obtained, and it is found that
34 people in the sample are users of foreign-made products; the
rest are users of domestic products. Give a 95% confidence
interval for the share of foreign products in this market.
6-28
Example
A marketing research firm wants to estimate the share that foreign

companies have in the American market for certain products. A random
sample of 100 consumers is obtained, and it is found that 34 people in the
sample are users of foreign-made products; the rest are users of domestic
products. Give a 95% confidence interval for the share of foreign products
in this market.
(0.34)(0.66)
pq
p 2 z n  0.34  100
 0.34  (1.96)(0.04737)
1.96  0.34  0.0928
 0.2472,0.4328
Thus, the firm may be 95% confident that foreign manufacturers
control anywhere from 24.72% to 43.28% of the market.
6-29
Confidence Interval or Interval Estimate for 

When  Is Unknown - The t Distribution
If the population standard deviation, , is not known,

replace
 with the sample standard deviation,
normal, the resulting statistic: X   s. If the population is
s t 
n
has a t distribution with (n - 1)
degrees of freedom.
• The t is a family of bell-shaped and symmetric Standard normal
distributions, one for each number of degree of
freedom. t, df = 20
• The expected value of t is 0.
• For df > 2, the variance of t is df/(df-2). This is t, df =
greater than 1, but approaches 1 as the number 10
of degrees of freedom increases. The t is flatter
and has fatter tails than does the standard
normal.
• The t distribution approaches a standard normal

as the number of degrees of freedom increases 
6-30
Confidence Intervals for  when  is

Unknown- The t Distribution
A (1-)100% confidence interval for  when  is not known

(assuming a normally distributed population) is given by:
s
x 
t 2 n
where t2 is the value of the t distribution with n-1 degrees of

freedom that cuts off a tail area of 2 to its right.
6-31
The t Distribution
df t0.100 t0.050 t0.025 t0.010 t0.005
--- ----- ----- ------ ------ ------
1 3.078 6.314 12.706 31.821 63.657 t D is trib utio n: d f= 1
2 1.886 2.920 4.303 6.965 9.925 0
3 1.638 2.353 3.182 4.541 5.841 0 .4
4 1.533 2.132 2.776 3.747 4.604
5 1.476 2.015 2.571 3.365 4.032
6 1.440 1.943 2.447 3.143 3.707 0 .3
7 1.415 1.895 2.365 2.998 3.499 Area = 0.10 Area = 0.10
8 1.397 1.860 2.306 2.896 3.355
}
f(t)
9 1.383 1.833 2.262 2.821 3.250 0 .2
10 1.372 1.812 2.228 2.764 3.169
11 1.363 1.796 2.201 2.718 3.106
0 .1
12 1.356 1.782 2.179 2.681 3.055
13 1.350 1.771 2.160 2.650 3.012
14 1.345 1.761 2.145 2.624 2.977
0 .0
15 1.341 1.753 2.131 2.602 2.947 -1.372 1.372
-2.228 0
16 1.337 1.746 2.120 2.583 2.921 2.228
}
17 1.333 1.740 2.110 2.567 2.898 t
18 1.330 1.734 2.101 2.552 2.878
19 1.328 1.729 2.093 2.539 2.861 Area = 0.025 Area = 0.025
20 1.325 1.725 2.086 2.528 2.845
21 1.323 1.721 2.080 2.518 2.831
22 1.321 1.717 2.074 2.508 2.819
23
24
1.319
1.318
1.714
1.711
2.069
2.064
2.500
2.492
2.807
2.797 Whenever  is not known (and the population is
25
26
1.316
1.315
1.708
1.706
2.060
2.056
2.485
2.479
2.787
2.779 assumed normal), the correct distribution to use is
27
28
1.314
1.313
1.703
1.701
2.052
2.048
2.473
2.467
2.771
2.763 the t distribution with n-1 degrees of freedom.
29
30
1.311
1.310
1.699
1.697
2.045
2.042
2.462
2.457
2.756
2.750 Note, however, that for large degrees of freedom,
40
60
1.303
1.296
1.684
1.671
2.021
2.000
2.423
2.390
2.704
2.660 the t distribution is approximated well by the Z
0
1.289
1.282
1.658
1.645
1.980
1.960
2.358
2.326
2.617
2.576 distribution.
6-32
Example
A stock market analyst wants to estimate the average return on a
certain stock. A random sample of 15 days yields an average
(annualized) return of 10.37% and a standard deviation of s =
3.5%. Assuming that the returns are normal, give a 95%
confidence interval for the average return on this stock.
6-33
Example
A stock market analyst wants to estimate the average return on a certain
stock. A random sample of 15 days yields an average (annualized) return
of and a standard deviation of s = 3.5%. Assuming the returns are normal,
give a 95% confidence interval for the average return on this stock.
x  10.37%
df t0.100 t0.050 t0.025 t0.010 t0.005
---
1
-----
3.078
-----
6.314
------
12.706
------
31.821
------
63.657
The critical value of t for df = (n -1) = (15
.
.
.
.
.
.
.
.
.
.
.
.
-1)
13
. .
1.350
.
1.771
.
2.160
.
2.650
.
3.012
t 0.025  area of 0.025 is:
=14 and a right-tail
14
15
1.345
1.341
1.761
1.753
2.145
2.131
2.624
2.602
2.977
2.947 The corresponding
2.145 confidences interval
.
.
.
.
.
.
.
.
.
.
.
. or
interval estimate is: x  t
. . . . . . 0.025 n
3.
 10.37 
515
2.145
 10.37 

1.94
8.43,12.31
Problem
The owner of a large equipment rental company wants to make a
rather quick estimate of the average number of days a piece of
ditchdigging equipment is rented out per person per time. The company
has records of all rentals, but the amount of time required to
conduct an audit of all accounts would be prohibitive. The owner
decides to take a random sample of rental invoices. Fourteen different
rentals of ditchdiggers are selected randomly from the files, yielding
the following data. She uses these data to construct a 99% confidence
interval to estimate the average number of days that a ditchdigger is
rented and assumes that the number of days per rental is normally
distributed in the population.
Solution
sample mean is 2.14 and the sample standard deviation is 1.29.

6-36
Confidence Intervals for the Population Variance:

The Chi-Square (2) Distribution
• The sample variance, s2, is an unbiased estimator of the
population variance, 2.
• The relationship of the sample variance to the
population variance is captured by chi-square (2)
distribution.
6-37
The Chi-Square (2) Distribution
Chi-Square D istrib ution: df=10 , df=3 0 ,

 The chi-square random variable df=50
0 .10
cannot be negative, so it is bound 0 .09

0 .08
df = 10
by zero on the left. 0 .07

0 .06
f()
df = 30
 The chi-square distribution is
2
0 .05
0 .04
df = 50
skewed to the right.
0 .03
0 .02
0 .01
 The chi-square distribution 0 .00

0 50 1 00
approaches a normal as the 2
Indegrees
ofsampling
freedom from a normal population, the random variable:
increase.
 2  ( n  1)s
2
2
has a chi - square distribution with (n - 1) degrees of
freedom.
6-38
Values and Probabilities of Chi-

Square Distributions
Area in Right Tail
.995 .990 .975 .950 .900 .100 .050 .025 .010

.005
Area in Left Tail
df .005 .010 .025 .050 .100 .900 .950 .975 .990 .995
1 0.0000393 0.000157 0.000982 0.000393 0.0158 2.71 3.84 5.02 6.63 7.88
2 0.0100 0.0201 0.0506 0.103 0.211 4.61 5.99 7.38 9.21 10.60
3 0.0717 0.115 0.216 0.352 0.584 6.25 7.81 9.35 11.34 12.84
4 0.207 0.297 0.484 0.711 1.06 7.78 9.49 11.14 13.28 14.86
5 0.412 0.554 0.831 1.15 1.61 9.24 11.07 12.83 15.09 16.75
6 0.676 0.872 1.24 1.64 2.20 10.64 12.59 14.45 16.81 18.55
7 0.989 1.24 1.69 2.17 2.83 12.02 14.07 16.01 18.48 20.28
8 1.34 1.65 2.18 2.73 3.49 13.36 15.51 17.53 20.09 21.95
9 1.73 2.09 2.70 3.33 4.17 14.68 16.92 19.02 21.67 23.59
10 2.16 2.56 3.25 3.94 4.87 15.99 18.31 20.48 23.21 25.19
11 2.60 3.05 3.82 4.57 5.58 17.28 19.68 21.92 24.72 26.76
12 3.07 3.57 4.40 5.23 6.30 18.55 21.03 23.34 26.22 28.30
13 3.57 4.11 5.01 5.89 7.04 19.81 22.36 24.74 27.69 29.82
14 4.07 4.66 5.63 6.57 7.79 21.06 23.68 26.12 29.14 31.32
15 4.60 5.23 6.26 7.26 8.55 22.31 25.00 27.49 30.58 32.80
16 5.14 5.81 6.91 7.96 9.31 23.54 26.30 28.85 32.00 34.27
17 5.70 6.41 7.56 8.67 10.09 24.77 27.59 30.19 33.41 35.72
18 6.26 7.01 8.23 9.39 10.86 25.99 28.87 31.53 34.81 37.16
19 6.84 7.63 8.91 10.12 11.65 27.20 30.14 32.85 36.19 38.58
20 7.43 8.26 9.59 10.85 12.44 28.41 31.41 34.17 37.57 40.00
21 8.03 8.90 10.28 11.59 13.24 29.62 32.67 35.48 38.93 41.40
22 8.64 9.54 10.98 12.34 14.04 30.81 33.92 36.78 40.29 42.80
23 9.26 10.20 11.69 13.09 14.85 32.01 35.17 38.08 41.64 44.18
24 9.89 10.86 12.40 13.85 15.66 33.20 36.42 39.36 42.98 45.56
25 10.52 11.52 13.12 14.61 16.47 34.38 37.65 40.65 44.31 46.93
26 11.16 12.20 13.84 15.38 17.29 35.56 38.89 41.92 45.64 48.29
27 11.81 12.88 14.57 16.15 18.11 36.74 40.11 43.19 46.96 49.65
28 12.46 13.56 15.31 16.93 18.94 37.92 41.34 44.46 48.28 50.99
29 13.12 14.26 16.05 17.71 19.77 39.09 42.56 45.72 49.59 52.34
30 13.79 14.95 16.79 18.49 20.60 40.26 43.77 46.98 50.89 53.67
6-39
Confidence Interval for the

Population Variance
A (1-)100% confidence interval for the population variance * (where the
population is assumed normal) is:
 2
(n  1)s 2
(n  1)s
 ,  2 

  1 
2
2 2
 2
where  is the value of the chi-square distribution with n - 1 degrees of freedom
2 
that cuts off an area to its right and  2  is the value of the distribution that
2
1
2 
cuts off an area of to its left (equivalently, an area of 1 to its
2 2
right).
* Note: Because the chi-square distribution is skewed, the confidence interval for the
population variance is not symmetric
Problem
The U.S. Bureau of Labor Statistics publishes data on the
hourly compensation costs for production workers in
manufacturing for various countries. The latest figures
published for Greece show that the average hourly wage
for a production worker in manufacturing is $16.10. Suppose
the business council of Greece wants to know how consistent
this figure is. They randomly select 25 production workers in
manufacturing from across the country and determine that the
standard deviation of hourly wages for such workers is $1.12.
Use this information to develop a 95% confidence interval to
estimate the population variance for the hourly wages of
production workers in manufacturing in Greece. Assume that
the hourly wages for production workers across the country in
manufacturing are normally distributed.
Solution
A 95% confidence means that alpha is 1 - .95 = .05. This
value is split to determine the area in each tail of the chi-
square distribution: α/2 = .025.
The confidence interval can be

determined
The business council can estimate with 95% confidence

that the population variance of the hourly wages of
production workers in manufacturing in Greece is
between 0.7648 and 2.4276.
6-42
Example
In an automated process, a machine fills cans of coffee. If the average
amount filled is different from what it should be, the machine may
be adjusted to correct the mean. If the variance of the filling
process is too high, however, the machine is out of control and needs
to be repaired. Therefore, from time to time regular checks of the
variance of the filling process are made. This is done by randomly
sampling filled cans, measuring their amounts, and computing the
sample variance. A random sample of 30 cans gives an estimate s2 =
18,540.
Give a 95% confidence interval for the population variance, 2.
6-43
Solution
In an automated process, a machine fills cans of coffee. If the average
amount filled is different from what it should be, the machine may
be adjusted to correct the mean. If the variance of the filling
process is too high, however, the machine is out of control and needs
to be repaired. Therefore, from time to time regular checks of the
variance of the filling process are made. This is done by randomly
sampling filled cans, measuring their amounts, and computing the
sample variance. A random sample of 30 cans gives an estimate s2 =
18,540.
Give a 95% confidence interval for the population variance, 2.
 2  
(n
  2 1)s 2
, (n  1)s    (30 45.7
1)18540, (30 
 

1)18540  2
 11765,33604

2  1 2
16.0
7-1
Hypothesis
Testing
7-2
Using Statistics
• A hypothesis is a statement or assertion about the state of nature
(about
the true value of an unknown population parameter):
 The accused is innocent
  = 100
• Every hypothesis implies its contradiction or alternative:
 The accused is guilty
  100
• A hypothesis is either true or false, and you may fail to reject it or
you may reject it on the basis of information:
 Trial testimony and evidence
 Sample data
7-3
Statistical Hypothesis Testing

• A null hypothesis, denoted by H0, is an assertion about one or more
population parameters. This is the assertion we hold to be true until we have
sufficient statistical evidence to conclude otherwise.
 Often represents the status quo situation or an existing belief.
There is nothing new happening, the old theory is still true, the old
standard is correct, and the system is in control
 Is maintained, or held to be true, until a test leads to its rejection in favor of
the alternative hypothesis.
Is accepted as true or rejected as false on the basis of a consideration of a
test statistic.
• The alternative hypothesis, denoted by H1 or Ha ,is the assertion of all

situations not covered by the null hypothesis.
 The new theory is true, the are new standards, the system is out of
control,
and/or something is happening
Generally speaking, new hypotheses that business researchers want
7-4
Statistical Hypothesis Testing
• H0 and H1 are:
 Mutually exclusive
– Only one can be true.
 Exhaustive
– Together they cover all possibilities, so one or the other must be
true.
Example
Suppose flour packaged by a manufacturer is sold by weight; and a
particular size of package is supposed to average 40 ounces.
Suppose the manufacturer wants to test to determine whether their
packaging process is out of control as determined by the weight of
the flour packages.
7-6
1-Tailed and 2-Tailed Tests
The tails of a statistical test are determined by the need for an action. If action
is to be taken if a parameter is greater than some value a, then the alternative
hypothesis is that the parameter is greater than a, and the test is a right-tailed
test. H0:   50
H1:   50
If action is to be taken if a parameter is less than some value a, then the

alternative hypothesis is that the parameter is less than a, and the test is a
left-
tailed test. H0:   50
H1 :  
50
If action is to be taken if a parameter is either greater than or less than some
value a, then the alternative hypothesis is that the parameter is not equal to
and the test is a two-tailed test.
a, H0:   50
H1 :  
50
Example
Suppose a company has held an 18% share of the market. However,
because of an increased marketing effort, company officials believe the
company’s market share is now greater than 18%, and the officials
would like to prove it.
Example
7-9
Hypothesis about other Parameters
• Hypotheses about other parameters such as population

proportions and population variances are also possible.
For example
 H0: p  40%
 H1: p < 40%
 H0:   
 H1:   
Example
A vendor claims that his company fills any accepted order, on the
average, in at most six working days. You suspect that the average is
greater than six working days and want to test the claim. How will you
set up the null and alternative hypotheses?
Example
A vendor claims that his company fills any accepted order, on the
average, in at most six working days. You suspect that the average is
greater than six working days and want to test the claim. How will you
set up the null and alternative hypotheses?
Example
A manufacturer of golf balls claims that the variance of the weights of
the company’s golf balls is controlled to within 0.0028 oz2. If you wish
to test this claim, how will you set up the null and alternative
hypotheses?
Example
A manufacturer of golf balls claims that the variance of the weights of
the company’s golf balls is controlled to within 0.0028 oz2. If you wish
to test this claim, how will you set up the null and alternative
hypotheses?
Example
At least 20% of the visitors to a particular commercial Web site where
an electronic product is sold are said to end up ordering the product. If
you wish to test this claim, how will you set up the null and alternative
hypotheses?
Example
Because the company officials are only interested in “proving” that the
market share has increased and the inclusion of the “less than” sign in the
null hypothesis is confusing. Also, If the equal part of the null hypothesis is
rejected because the market share is seemingly greater, then certainly the
“less than” portion of the null hypothesis is also rejected because it is
further away from “greater than” than is “equal.” Using this logic, the null
hypothesis for the market share problem can be written as
Hypothesis Testing Process
The first four steps in testing

hypotheses should always
be completed before the
study is undertaken
Example
Suppose flour packaged by a manufacturer is sold by weight; and a
particular size of package is supposed to average 40 ounces.
Suppose the manufacturer wants to test to determine whether their
packaging process is out of control as determined by the weight of
the flour packages.
Solution
i) Suppose a sample of 100 such packages is randomly selected, and
a sample mean of 40.01 ounces is obtained.
ii) Suppose a sample mean of 50 ounces is obtained for 100 packages.
When is the sample mean so far away from the population mean that
the null hypothesis is rejected?
• Confidence Interval
7-19
Rejection Region
• The rejection region of a statistical hypothesis test is the range of

numbers that will lead us to reject the null hypothesis in case the
test statistic falls within this range. The rejection region, also
called the critical region, is defined by the critical points. The
rejection region is defined so that, before the sampling takes
place, our test statistic will have a probability  of falling within
the rejection region if the null hypothesis is true.
7-20
Nonrejection Region
• The nonrejection region is the range of values (also
determined by the critical points) that will lead us not to reject
the null hypothesis if the test statistic should fall within this
region. The nonrejection region is designed so that, before
the sampling takes place, our test statistic will have a
probability 1-
 of falling within the nonrejection region if the null
hypothesis is true
In a two-tailed test, the rejection region consists of the
values in both tails of the sampling distribution.
7-21
The Concepts of Hypothesis Testing

• A test statistic is a sample statistic computed from sample data.
The value of the test statistic is used in determining whether or not
we may reject the null hypothesis.
• The decision rule of a statistical hypothesis test is a rule that
specifies the conditions under which the null hypothesis may be
rejected.
Consider H0:  = 100. We may have a decision rule that says:

“Reject H0 if the sample mean is less than 95 or more than
105.”
7-22
Decision Making
• There are two possible states of nature:

H0 is true
H0 is false
• There are two possible decisions:
Fail to reject H0 as true
Reject H0 as false
7-23
Decision Making
• A decision may be correct in two ways:
Fail to reject a true H0
Reject a false H0
• A decision may be incorrect in two ways:
Type I Error: Reject a true H0
• The Probability of a Type I error is denoted by .
Type II Error: Fail to reject a false H0
• The Probability of a Type II error is denoted by .
7-24
Errors in Hypothesis Testing

• A decision may be incorrect in two ways:
 Type I Error: Reject a true H0
◼ T h e Probability of a Type I error is denoted by
.
  is called the level of significance of the test
 Type II Error: Accept a false H0
◼ T h e Probability of a Type II error is denoted by
.
 1 -  is called the power of the test.
 = P(Reject H 0 H 0 is
•  true)
  =areP(Accept
and conditional probabilities:
H0 H 0 is false)
7-25
Type I and Type II Errors

A contingency table illustrates the possible outcomes
of a statistical hypothesis test.
The “state of nature” is how things actually are and the “action” is
the decision that the business researcher actually makes
Type - I Error
Suppose the flour-packaging process actually is “in control” and is
averaging 40 ounces of flour per package. Suppose also that a business
researcher randomly selects 100 packages, weighs the contents of
each, and computes a sample mean. It is possible, by chance, to
randomly select 100 of the more extreme packages (mostly heavy
weighted or mostly light weighted) resulting in a mean that falls in the
rejection region. The decision is to reject the null hypothesis even
though the population mean is actually 40 ounces. In this case, the
business researcher has committed a Type I error.
 if a manager fires an employee because some evidence indicates
that she is stealing from the company and if she really is not stealing
from the company,
 Suppose a worker on the assembly line of a large manufacturer
hears an unusual sound and decides to shut the line down. If the
sound turns out not to be related to the assembly line and no
problems are occurring with the assembly line
7-29
Statistical Significance
While the null hypothesis is maintained to be true throughout a
hypothesis test, until sample data lead to a rejection, the aim of a
hypothesis test is often to disprove the null hypothesis in favor of
the alternative hypothesis. This is because we can determine and
regulate , the probability of a Type I error, making it as small as we
desire, such as 0.01 or 0.05. Thus, when we reject a null hypothesis,
we have a high level of confidence in our decision, since we know
there is a small probability that we have made an error.
Example
A survey of CPAs across the United States found that the average
net income for sole proprietor CPAs is $74,914.Because this survey
is now more than ten years old, an accounting researcher wants to
test this figure by taking a random sample of 112 sole proprietor
accountants in the United States which showed a sample mean of
$78,695. Assume the population standard deviation of net incomes
for sole proprietor CPAs is $14,530.
Solution
Step 1
Step 2
Solution
Step 3
Type I error rate, or alpha, which is .05 in this problem
Step 4
Because the test is two tailed and alpha is .05, there is 2 or .025 area in each
of the tails of the distribution. Thus, the rejection region is in the two ends
of the distribution with 2.5% of the area in each.
Solution
Step 5
Step 6
Because this test statistic, z = 2.75, is greater than the critical value of z in the
upper tail of the distribution, z = +1.96,
Step 7
Reject the null hypothesis
Step 8
Statistically, the
researcher has enough
7-34
Critical Value Method

A company that delivers packages within a large metropolitan
area claims that it takes an average of 28 minutes for a package to
be delivered from your door to the destination. Suppose you
want to carry out a hypothesis test of this claim at 95%
confidence by taking a sample of 100 packages, with an average
delivery time of
31.5 minutes & standard deviation of 5 minutes.
7-35
Picturing Hypothesis Testing
95% confidence
Population interval around
mean under H0 observed sample mean
 = 28 30.52 x = 31.5 32.48

It seems reasonable to reject the null hypothesis, H0:  = 28, since the
hypothesized value lies outside the 95% confidence interval. If we are 95% sure
that the population mean is between 30.52 and 32.58 minutes, it is very unlikely
that the population mean will actually be 28 minutes.
Note that the population mean may be 28 (the null hypothesis might be true), but
then the observed sample mean, 31.5, would be a very unlikely occurrence. There
is still the small chance ( = 0.05) that we might reject the true null hypothesis.
 represents the level of significance of the test.
7-36
Nonrejection Region
If the observed sample mean falls within the nonrejection region, then you fail
to reject the null hypothesis as true. Construct a 95% nonrejection region
around the hypothesized population mean, and compare it with the 95%
confidence interval around the observed sample mean:
s 5 s 5
 0 z  28  95% non- 95% Confidence x  z.025  31.5 
.025
n 1.96 100 rejection region Interval n 1.96 100
around the around the
population Mean Sample Mean
 28.98  27,02,28.98  31.5.98  30.52,32.48
27.02 0=28 28.98 30.52 x

32.48
The nonrejection region and the confidence interval are the same width, but
centered on different points. In this instance, the nonrejection region does not
include the observed sample mean, and the confidence interval does not
include the hypothesized population mean.
7-37
Solution
A company that delivers packages within a large metropolitan
area claims that it takes an average of 28 minutes for a package to
be delivered from your door to the destination. Suppose you
want to carry out a hypothesis test of this claim at 95%
confidence by taking a sample of 100 packages, with an average
delivery time of
31.5 minutes & standard deviation of 5 minutes. 5
Set the null and alternative hypotheses: s
H0:  = 28 x  z .025
n  31.5  1.96
H1:   28 100
Collect sample data:  31.5  .98  30.52, 32.48

n = 100
We can be 95% sure that the average
x = 31.5
time for all packages is between
s=5
30.52 and 32.48 minutes.
Since the asserted value, 28 minutes,
Construct a 95% confidence interval
is not in this 95% confidence interval,
for the average delivery times of all
we may reasonably reject the null
packages:
Example
Solution
7-40
The p-Value
The p-value is the probability of obtaining a value of the test statistic as

extreme as, or more extreme than, the actual value obtained, when the null
hypothesis is true.
The p-value is the smallest level of significance, , at which the null

hypothesis may be rejected using the obtained value of the test statistic.
RULE: When the p-value is less than  , reject H0.

7-41
The p-Value
7-42
Example
An automatic bottling machine fills cola into two liter (2000 cc) bottles. A
consumer advocate wants to test the null hypothesis that the average amount
filled by the machine into a bottle is at least 2000 cc. A random sample of 40
bottles coming out of the machine was selected and the exact content of the
selected bottles are recorded. The sample mean was 1999.6 cc. The population
standard deviation is known from past experience to be 1.30 cc. Test this
hypotheses at 95% confidence with the help of p-value.
7-43
Example
n = 40
H0:   2000 x=
H 1:   1999.6
2000 n = 40  = 1.3
For  =
0.05, the x 1999.6 -
z 0
critical value 2000
x  0
The
of testz statistic
is is: z    =
1.3
-1.645 n
n 40
Do not reject H0 if: [z -
=  1.95  Reject
1.645] H 0
7-44
Example
z  x  0 = 1999.6-
1.3
2000
H0:   2000
n 40
H 1:  
2000
= 1.95
n = 40, 0 =
p - value  P(Z  -1.95)
2000, x-bar  0.5000-0.4744
The test statistic is: x  0
= 1999.6, z    0.0256
 = 1.3 n
Example
In an attempt to determine why customer service is important to

managers in the United Kingdom, researchers surveyed managing
directors of manufacturing plants in Scotland. One of the reasons
proposed was that customer service is a means of retaining customers.
On a scale from 1 to 5, with 1 being low and 5 being high, the survey
respondents rated this reason more highly than any of the others, with
a mean response of 4.30. Suppose U.S. researchers believe American
manufacturing managers would not rate this reason as highly and
conduct a hypothesis test to prove their theory. Alpha is set at .05. Data
are gathered and the following results are obtained. Use these data to
determine whether U.S. managers rate this reason significantly lower
than the 4.30 mean ascertained in the United Kingdom. Assume from
previous studies that the population standard deviation is 0.574.
Solution
Step 1
Step 2
Step 3
Type I error rate, or alpha, which is .05 in this problem
Solution
Step 4
Because the test is one tailed test and alpha is .05, there is .05 area in the
left tail of the distribution.
Step 5,6
I)
II)
Solution
Step 7
1. Observed value method
Because the observed test statistic is not less than the critical value and is
not in the rejection region, the statistical conclusion is that the null
hypothesis cannot be rejected
2. Critical Value method
Because the mean obtained from the sample data is 4.156, the researchers
fail to reject the null hypothesis
3. p-Value method
The observed test statistic is z = -1.42. The probability of getting a z value at
least this extreme when the null hypothesis is true is .5000 - .4222 = .0778.
Step 8
The test does not result in enough evidence to conclude that U.S. managers
think it is less important to use customer service as a means of retaining
customers than do UK managers. Customer service is an important tool for
retaining customers in both countries according to managers.
7-49
7-62
The Hypothesis Test
We will see the three different types of hypothesis tests, namely
 Tests of hypotheses about population means.

 Tests of hypotheses about population proportions.
 Tests of hypotheses about population variances.
7-63
Testing Population Means
• Cases in which the test statistic is Z
  is known and the population is normal.

 is known and the sample size is at least 30. (The
population need not be normal)
The formula for calculating Z is :
z
x
 
 
n
7-64
Testing Population Means
• Cases in which the test statistic is t
 is unknown but the sample standard deviation is known and

the population is normal.
The formula for calculating t is :
t
x
 s 
 
n
7-65
Example
A coin is to tested for fairness. It is tossed 25 times and only 8 Heads are
observed. Test if the coin is fair at an  of 5% (significance level).
7-66
Example
A coin is to tested for fairness. It is tossed 25 times and only 8 Heads are
observed. Test if the coin is fair at an  of 5% (significance level).
Let p denote the probability of a

Head H0: p  0.5
H1: p  
Because this is a 2-tailed test, the p-
value = 2*P(X  
From the binomial tables, with n = 25, p = 0.5, this
value 2*0.054 = 0.108.
Since 0.108 >  = 0.05,
then do not reject H0
7-67
Testing Population Variances
•For testing hypotheses about population variances, the test

statistic (chi-square) is:
 2
n
1s 2
0
2
where  is the claimed value of the population variance in

2
0
null hypothesis. The degrees of freedom for this chi-square

the
random variable is (n – 1).
Note: Since the chi-square table only provides the critical values, it
cannot be used to calculate exact p-values. As in the case of the t-tables,
only a range of possible values can be inferred.
7-68
Example
A manufacturer of golf balls claims that they control the weights of the golf balls
accurately so that the variance of the weights is not more than 1 mg2. A random sample
of 31 golf balls yields a sample variance of 1.62 mg2. Is that sufficient evidence to
reject the claim at an  of 5%?
7-69
Solution
A manufacturer of golf balls claims that they control the weights of the golf balls
accurately so that the variance of the weights is not more than 1 mg2. A random sample
of 31 golf balls yields a sample variance of 1.62 mg2. Is that sufficient evidence to
reject the claim at an  of 5%?
Let 2 denote the population variance. Then

H0: 2  1
H1: 2  
The p-value of 0.0173
Since this value is less than the  of 5%, we
reject the null hypothesis.
7-70
Example
As part of a survey to determine the extent of required in-cabin storage capacity, a

researcher needs to test the null hypothesis that the average weight of carry-on baggage
per person is   = 12 pounds, versus the alternative hypothesis that the average weight is
not 12 pounds. The analyst wants to test the null hypothesis at  = 0.05.
7-71
Example
As part of a survey to determine the extent of required in-cabin storage capacity, a

researcher needs to test the null hypothesis that the average weight of carry-on baggage
per person is   = 12 pounds, versus the alternative hypothesis that the average weight is
not 12 pounds. The analyst wants to test the null hypothesis at  = 0.05.

H0:  = 12
H1:   12
0.8
0.7 .95
0.6
0.5
For  = 0.05, critical values of z are 0.4
±1.96 x
0.3
.025 .025
The test statistic is: z 0 s

0.2
0.1
0.0
n -1.96  1.96 z
Do not reject H0 if: [-1.96  z 1.96]

Lower Rejection Nonrejection Upper Rejection
Region Region Region
Reject H0 if: [z <-1.96] or z 1.96]
7-72
Additional Examples (a): Solution
n = 144 The Standard Normal Distribution

0.8
0.7 .95
x = 14.6 0.6
0.5
0.4
s = 7.8 0.3
0.2
.025
.025
0.1
0.0
x = 14.6-12
z 0 s
z
7.8 -1.96  1.96
n 144 Lower Rejection

Region
Nonrejection
Region
Upper Rejection
Region
= 2.6  4
0.65
Since the test statistic falls in the upper rejection region, H0 is rejected, and we may
conclude that the average amount of carry-on baggage is more than 12 pounds.
7-73
Examples
An insurance company believes that, over the last few years, the average liability
insurance per board seat in companies defined as “small companies” has been
$2000. Using  = 0.01, test this hypothesis using Growth Resources, Inc. survey
data.
7-74
Examples
An insurance company believes that, over the last few years, the average liability
insurance per board seat in companies defined as “small companies” has been
$2000. Using  = 0.01, test this hypothesis using Growth Resources, Inc. survey
data.
n = 100
H0:  = 2000 x = 2700
H1:   2000 s = 947
For  = 0.01, critical values of z are ±2.576 x0 2700 - 2000

z =
x 947
The test statistic is: 
z 0 s
s
100
n n
700
=  7.39  Reject H
Do not reject H0 if: [-2.576  z  2.576] 94.7 0
Reject H0 if: [z <-2.576] or z 2.576]

7-75
Example

Since the test statistic falls in
0.8
0.7 .99 the upper rejection region, H0
0.6
0.5
0.4
is rejected, and we may
0.3
0.2
.005 .005
conclude that the average
0.1
0.0
insurance liability per board
-2.576  2.576 z seat in “small companies” is

more than $2000.
Lower Rejection Nonrejection Upper Rejection
7-76
Example
The average time it takes a computer to perform a certain task is believed to be 3.24
seconds. It was decided to test the statistical hypothesis that the average performance
time of the task using the new algorithm is the same, against the alternative that the
average performance time is no longer the same, at the 0.05 level of significance.
7-77
Examples
The average time it takes a computer to perform a certain task is believed to be 3.24
seconds. It was decided to test the statistical hypothesis that the average performance
time of the task using the new algorithm is the same, against the alternative that the
average performance time is no longer the same, at the 0.05 level of significance.
H0:  = 3.24 n = 200

H 1:   x = 3.48
3.24 s = 2.8
For  = x
x 0 3.48 -
z
0.05,test statistic is: z  s
The
0
3.24
=
critical n s
values of z 2.8
Do not reject H0 if: [-1.96  z 1.96] 0.24  1.21
are ±1.96 = n
0.20
 Do not reject H
0
200
Reject H0 if: [z < -1.96] or z 1.96]
7-78
Example

0.8 Since the test statistic falls in
0.7
the nonrejection region, H0
.95
0.6
0.5
0.4
0.3
is not rejected, and we may
0.2 .025
.025
0.1
performance time has not

0.0
z
-1.96 
1.96

changed from 3.24 seconds.
Nonrejection
Lower Rejection Region Upper Rejection
Region Region
7-79
Example
According to the Japanese National Land Agency, average land prices in central Tokyo
soared 49% in the first six months of 1995. An international real estate investment
company wants to test this claim against the alternative that the average price did not
rise by 49%, at a 0.01 level of significance.
7-80
Example
According to the Japanese National Land Agency, average land prices in central Tokyo
soared 49% in the first six months of 1995. An international real estate investment
company wants to test this claim against the alternative that the average price did not
rise by 49%, at a 0.01 level of significance.
H0:  = 49 n = 18
H 1:   x = 38
s = 14
49 n = 18
For  = 0.01 and (18-1) = 17 x   38 - 49
0
df , critical values of t are t 
s
=
±2.898 x 

14
The test statistic is: t  0
s n
n
- 1 11 8
=  3.33  Reject H
Do not reject H0 if: [-2.898  t  2.898] 3.3 0
Reject H0 if: [t < -2.898] or t 

2.898]
7-81
Example
The t Distribution Since the test statistic falls in

0.8
0.7 .99
the rejection region, H0
0.6
0.5
is
rejected, and we may
0.4
0.3
conclude
0.2
0.1
.005 .005
that the average price has
0.0
not risen by 49%. Since the
t
test statistic is in the lower

-2.898  2.898

Lower Rejection
Region
Nonrejection
Region
Upper Rejection
Region
rejection region, we may
price has risen by less than
49%.
7-82
Example
Canon, Inc,. has introduced a copying machine that features two-color copying capability
in a compact system copier. The average speed of the standard compact system copier is
27 copies per minute. Suppose the company wants to test whether the new copier has the
same average speed as its standard compact copier. Conduct a test at an  = 0.05 level of
significance.
7-83
Example
Canon, Inc,. has introduced a copying machine that features two-color copying capability
in a compact system copier. The average speed of the standard compact system copier is
27 copies per minute. Suppose the company wants to test whether the new copier has the
same average speed as its standard compact copier. Conduct a test at an  = 0.05 level of
significance.
n = 24
H0:  = 27 x = 24.6
H1:   27 s = 7.4
n = 24
For  = 0.05 and (24-1) = 23 x 0 24.6 -
t
df , critical values of t are 27 =
±2.069 x s
 7.4
The test statistic is: t  s
0
n =
-2.4
n  1.59  Do not reject H
1.51 0
Do not reject H0 if: [-2.069
 t  2.069] 24
Reject H0 if: [t < -2.069] or
7-84
Example
The t Distribution
0.8 Since the test statistic falls in
0.7
the nonrejection region, H0
.95
0.6
0.5
0.4
0.3
is not rejected, and we may
.025 .025
0.2
0.1
not conclude that the
average speed is different
0.0
t
-2.069 
2.069
 from 27 copies per minute.
Nonrejection
Lower Rejection Region Upper Rejection
Region Region
7-85
Example
An investment analyst for Goldman Sachs and Company wanted to test the hypothesis
made by British securities experts that 70% of all foreign investors in the British market
were American. The analyst gathered a random sample of 210 accounts of foreign
investors in London and found that 130 were owned by U.S. citizens. At the  = 0.05
level of significance, is there evidence to reject the claim of the British securities
experts?
7-86
Example
An investment analyst for Goldman Sachs and Company wanted to test the hypothesis
made by British securities experts that 70% of all foreign investors in the British market
were American. The analyst gathered a random sample of 210 accounts of foreign
investors in London and found that 130 were owned by U.S. citizens. At the  = 0.05
level of significance, is there evidence to reject the claim of the British securities
experts?
n = 210
H0: p = 0.70 130
H1: p  0.70 p  0.619
= 210
n = 210
For  = 0.05 critical values of z are ±1.96 p -
0 0.619 - 0.70
z= p
The test statistic is: p  p (0.70)(0.30)
z p0 q 0
0
= 210
n pq
00
-0.081
Do not reject H0 if: [-1.96  = n  2.5614  Reject H
0.0316
zReject H0 if: [z < -1.96] or z  1.96]
 1.96]
0
7-87
Example
The EPA sets limits on the concentrations of pollutants emitted by various industries. Suppose that the
upper allowable limit on the emission of vinyl chloride is set at an average of 55 ppm within a range of two
miles around the plant emitting this chemical. To check compliance with this rule, the EPA collects a
random sample of 100 readings at different times and dates within the two-mile range around the plant. The
findings are that the sample average concentration is 60 ppm and the sample standard deviation is 20 ppm.
Is there evidence to conclude that the plant in question is violating the law?
7-88
Example
The EPA sets limits on the concentrations of pollutants emitted by various industries. Suppose that the
upper allowable limit on the emission of vinyl chloride is set at an average of 55 ppm within a range of two
miles around the plant emitting this chemical. To check compliance with this rule, the EPA collects a
random sample of 100 readings at different times and dates within the two-mile range around the plant. The
findings are that the sample average concentration is 60 ppm and the sample standard deviation is 20 ppm.
Is there evidence to conclude that the plant in question is violating the law?
n = 100
H0:   55
x = 60
H1:  55 s = 20
n = 100
For  = 0.01, the critical x 0 60 - 55
z =
value of z is 2.326 s
20
x0
The test statistic is: z s n
5  2.5
2 100
n =  Reject H
0
Do not reject H0 if: [z  2.326]
Reject H0 if: z 2.326]
7-89
Example
Critical Point for a Right-Tailed Test

Since the test statistic falls
in
the rejection region, H0 is
0 .4
0.99
rejected, and we may
0 .3
f(z)
0 .2
conclude
0 .1

that the average concentration
0 .0
-5 0 5
of vinyl chloride is more
than 55 ppm.
z 2.326
2.5
Nonrejection Rejection
Region Region
7-90
Example
A certain kind of packaged food bears the following statement on the package: “Average net weight 12 oz.”
Suppose that a consumer group has been receiving complaints from users of the product who believe that they are
getting smaller quantities than the manufacturer states on the package. The consumer group wants, therefore, to
test the hypothesis that the average net weight of the product in question is 12 oz. versus the alternative that the
packages are, on average, underfilled. A random sample of 144 packages of the food product is collected, and it
is found that the average net weight in the sample is 11.8 oz. and the sample standard deviation is 6 oz. Given
these findings, is there evidence the manufacturer is underfilling the packages?
7-91
Example
A certain kind of packaged food bears the following statement on the package: “Average net weight 12 oz.”
Suppose that a consumer group has been receiving complaints from users of the product who believe that they are
getting smaller quantities than the manufacturer states on the package. The consumer group wants, therefore, to
test the hypothesis that the average net weight of the product in question is 12 oz. versus the alternative that the
packages are, on average, underfilled. A random sample of 144 packages of the food product is collected, and it
is found that the average net weight in the sample is 11.8 oz. and the sample standard deviation is 6 oz. Given
these findings, is there evidence the manufacturer is underfilling the packages?
n = 144
H0:   12
x = 11.8
H1:   12
s=6
n = 144
For  = 0.05, the critical value
of z is -1.645 z
x  0 11.8 -12
x =
0 s 6
The test statistic is: z  s n 144
n
-.2
Do not reject H0 if: [z -1.645] =  0.4  Do not reject H
.5
Reject H0 if: z ]
0
7-92
Example
Critical Point for a Left-Tailed Test

0.4
the nonrejection region, H0 is

0.3 0.95
not rejected, and we may not
f(z)
0.2
 conclude that the manufacturer

0.1
is underfilling packages on
0.0
-5 0 average.
5
-0.4
-1.645 z
Rejection Nonrejection
Region Region
7-93
Additional Examples (i)
A floodlight is said to last an average of 65 hours. A competitor believes that the average life of the
floodlight is less than that stated by the manufacturer and sets out to prove that the manufacturer’s
claim is false. A random sample of 21 floodlight elements is chosen and shows that the sample
average is 62.5 hours and the sample standard deviation is 3. Using =0.01, determine whether
there is evidence to conclude that the manufacturer’s claim is false.
7-94
Additional Examples (i)
A floodlight is said to last an average of 65 hours. A competitor believes that the average life of the
floodlight is less than that stated by the manufacturer and sets out to prove that the manufacturer’s
claim is false. A random sample of 21 floodlight elements is chosen and shows that the sample
average is 62.5 hours and the sample standard deviation is 3. Using =0.01, determine whether
there is evidence to conclude that the manufacturer’s claim is false.
H0:   65
H 1:  
65 n = 21
For  =
0.01 an
(21-1) =
20 df, the
critical
value
-2.528
7-95
Additional Examples (i) : Continued
Critical Point for a Left-Tailed Test

0 .4
the rejection region, H0 is

0 .3 0.95
rejected, and we may conclude
f(t)
0 .2
 that the manufacturer’s claim

0 .1
is false, that the average

0 .0
-5 0
floodlight life is less than 65
5
-3.82
-2.528 t hours.
Rejection Nonrejection
Region Region
7-96
Additional Examples (j)

“After looking at 1349 hotels nationwide, we’ve found 13 that meet our standards.” This statement by the Small
Luxury Hotels Association implies that the proportion of all hotels in the United States that meet the association’s
standards is 13/1349=0.0096. The management of a hotel that was denied acceptance to the association wanted
to prove that the standards are not as stringent as claimed and that, in fact, the proportion of all hotels in the
United States that would qualify is higher than 0.0096. The management hired an independent research agency,
which visited a random sample of 600 hotels nationwide and found that 7 of them satisfied the exact standards
set by the association. Is there evidence to conclude that the population proportion of all hotels in the country
satisfying the standards set by the Small Luxury hotels Association is greater than 0.0096?
7-97
Additional Examples (j)

“After looking at 1349 hotels nationwide, we’ve found 13 that meet our standards.” This statement by the Small
Luxury Hotels Association implies that the proportion of all hotels in the United States that meet the association’s
standards is 13/1349=0.0096. The management of a hotel that was denied acceptance to the association wanted
to prove that the standards are not as stringent as claimed and that, in fact, the proportion of all hotels in the
United States that would qualify is higher than 0.0096. The management hired an independent research agency,
which visited a random sample of 600 hotels nationwide and found that 7 of them satisfied the exact standards
set by the association. Is there evidence to conclude that the population proportion of all hotels in the country
satisfying the standards set by the Small Luxury hotels Association is greater than 0.0096?
H0: p  0.0096
H1: p 
0.0096 n =
600
For  = 0.10 the critical value 1.282
The test statistic is:
Do not reject H0 if: [z 1.282]

7-98
Additional Examples (j) : Continued
Critical Point for a Right-Tailed Test

0 .4
the nonrejection region, H0 is

0. 0.90
3
not rejected, and we may not
f(z)
0.
2
conclude that proportion of
0. 
1
all hotels in the country that
0.
0
-5 0
z 1.282
5 meet the association’s
0.519 standards is greater than
Nonrejection
Region
Rejection
Region 0.0096.
7-99
The p-Value Revisited
Standard Normal Distribution Standard Normal Distribution
0.4 0.4
p-value=area to
p-value=area to
0.3 right of the test statistic 0.3
right of the test statistic
=0.3018
=0.0062
f(z
f(z
0.2 0.2
)
)
0.1 0.1
0.0 0.0
-5 0 0.519 5 -5 0 5
z 2.5 z
The p-value is the probability of obtaining a value of the test statistic as extreme as,
or more extreme than, the actual value obtained, when the null hypothesis is true.
The p-value is the smallest level of significance, , at which the null hypothesis
may be rejected using the obtained value of the test statistic.
7-100
The p-Value: Rules of Thumb
When the p-value is smaller than 0.01, the result is considered to

be very significant.
When the p-value is between 0.01 and 0.05, the result is

considered to be significant.
When the p-value is between 0.05 and 0.10, the result is

considered by some as marginally significant (and by most as not
significant).
When the p-value is greater than 0.10, the result is considered

not
significant.
7-101
p-Value: Two-Tailed Tests
p-value=double the area to

left of the test statistic
=2(0.3446)=0.6892
0.4
0.3
f(z
0.2
)
0.1
0.0
-5 0 5
-0.4 0.4
z
In a two-tailed test, we find the p-value by doubling the area

in the tail of the distribution beyond the value of the test
statistic.
7-102
The p-Value and Hypothesis Testing
The further away in the tail of the distribution the test statistic falls, the smaller
is the p-value and, hence, the more convinced we are that the null hypothesis
is false and should be rejected.
In a right-tailed test, the p-value is the area to the right of the test statistic if the
test statistic is positive.
In a left-tailed test, the p-value is the area to the left of the test statistic if
the test statistic is negative.
In a two-tailed test, the p-value is twice the area to the right of a positive
test statistic or to the left of a negative test statistic.
For a given level of significance, :

Reject the null hypothesis if and only if   p-value
8-1
The Comparison of Two

Populations
8-2
Using Statistics
• Inferences about differences between parameters of two
populations
 Paired-Observations
 Observe the same group of persons or things
 At two different times: “before” and “after”
 Under two different sets of circumstances or

“treatments”
 Independent Samples
 Observe different groups of persons or things
 At different times or under different sets of

circumstances
8-3
Paired-Observation Comparisons
• Population parameters may differ at two different times or

under two different sets of circumstances or treatments
because:
 The circumstances differ between times or treatments
 The people or things in the different groups are
themselves different
• By looking at paired-observations, we are able to minimize
the “between group” , extraneous variation.
8-4
Paired-Observation Comparisons
of Means
Test statistic for the paired- observations test
D  D
t df  n 1
S 0,
D
n
D  sample average for the differences
S D  sample standard deviation for the differences
n  sample size
 D  mean of the population of differences under the null
hypothesis
0
An assumption for this test is that the differences of the two populations are
normally distributed. The after measurement is not independent of the before
8-5
Example
A random sample of 16 viewers of Home Shopping Network was
selected for an experiment. All viewers in the sample had
recorded the amount of money they spent shopping during the
holiday season of the previous year. The next year, these people
were given access to the cable network and were asked to keep a
record of their total purchases during the holiday season. Home
Shopping Network managers want to test the null hypothesis that
their service does not increase shopping volume, versus the
alternative hypothesis that it does.
8-6
Example
8-7
Example
8-8
Example
H0: D  0
H1: D >
0
df = (n-1)
= (16-1) = D D
Test
15 Statistic: 0
t 
sD
n
D̅ = $32.81 , sD = $55.75, μD =
0
Critical Value: t0.05 = 1.753
Do not reject H0 if : t  1.753

Reject H0 if: t > 1.753
8-9
Solution
D  D 32.81  0
t  0   2.354
5 5.7 5
sD
n 16
t D is t r ib u t io n : d f = 1
5
0.
4
0.
3
f(t)
0.
2 Nonrejection Rejection
Region Region
0.
1
0.
0
-5 0 1.753 5 t
= t0.05
2.131
2.602
= t0.025
= t0.01
2.354=
test
statistic
Example
Suppose a stock market investor is interested in determining whether
there is a significant difference in the P/E (price to earnings) ratio for
companies from one year to the next. In an effort to study this
question, the investor randomly samples nine companies from the
Handbook of Common Stocks and records the P/E ratios for each of
these companies at the end of year 1 and at the end of year 2. Test at
α = 0.01
Solution
There is not enough evidence from the data to declare a significant difference in
the average P/E ratio between year 1 and year 2
Example
Consumers are asked to rate a company both before and
after viewing a video on the company twice a day for a
week. The data is displayed as follows. Use an alpha of .
05 to test to determine whether there is a significant
increase in the ratings of the company after the one-
week video treatment.
Assume that differences in ratings are normally distributed
in
the population.
Solution
The decision is to reject the null

8-14
Confidence Intervals for Paired

Observations
A (1-) 100% confidence interval for the mean difference

D:
D t sD
2
n
where t is the value of the t distribution with (n-1) degrees of
2
freedom that cuts off an area of  2 to its right,When the sample

size is large, we may approximate t 2 with z 2.
8-15
A Test for the Difference between Two Population

Means Using Independent Random Samples
• When paired data cannot be obtained, use independent

random samples drawn at different times or under different
circumstances.
 Large sample test if:
 Both n1 30 and n2 30 (Central Limit Theorem),
or
 Both populations are normal and 1 and 2 are both
known
 Small sample test if:
 Both populations are normal and 1 and 2 are
unknown
8-16
Comparisons of Two Population

Means: Testing Situations
• I: Difference between two population means is 0
 1= 2
◼ H0: 1 -2 = 0
◼ H1: 1 -2  0
• II: Difference between two population means is less than 0
 1 2
◼ H0: 1 -2  0
◼ H1: 1 -2  0
• III: Difference between two population means is less than
D
 1  2+D
◼ H0: 1 -2  D
◼ H1: 1 -2  D
8-17
Comparisons of Two Population

Means: Testing Situations
• IV: Difference between two population means is greater
than 0
 1 2
◼ H0: 1 -2  0
◼ H1: 1 -2  0
• V: Difference between two population means is
greater
than D
 1  2+ D
◼ H0: 1 -2  D
◼ H1: 1 -2  D
Test Statistic
Test Statistic
Test Statistic
Example
A random sample of 32 advertising managers from across the United States is
taken. The advertising managers are contacted by telephone and asked what their
annual salary is. A similar random sample is taken of 34 auditing managers. The
resulting salary data is listed below. The analyst wants to test at α= 0.05 whether
there is a difference in the average wage of an advertising manager and an
auditing manager.
Solution
The business researcher rejects the null hypothesis and can say that there
is significant difference between the average annual wage of an advertising
manager and the average annual wage of an auditing manager
Example
A sample of 87 professional working women showed that
the average amount paid annually into a private pension
fund per person was $3352. The population standard
deviation is
$1100. A sample of 76 professional working men showed
that the average amount paid annually into a private
pension fund per person was $5727, with a population
standard deviation of $1700. A women’s activist group
wants to “prove” that women do not pay as much per
year as men into private pension funds. If they use α= .01
and these sample data, will they be able to reject a null
hypothesis that women annually pay the same as or
more than men into private pension funds? Use the
eight-step hypothesis-testing process.
Solution
Null hypothesis is that women pay the same as or more than
men The alternative hypothesis
The evidence is substantial that women, on average, pay less than

men into private pension funds annually
Example
Until a few years ago, the market for consumer credit was
considered to be segmented. Higher-income, higher-spending
people tended to be American Express cardholders, and lower-
income, lower-spending people were usually Visa cardholders. In
the last few years, Visa has intensified its efforts to break into the
higher-income segments of the market by using magazine and
television advertising to create a high class image. Recently, a
consulting firm was hired by Visa to determine whether average
monthly charges on the American Express Gold Card are
approximately equal to the average monthly charges on Preferred
Visa. A random sample of 1,200 Preferred Visa cardholders was
selected, and the sample average monthly charge was found to
be $452. An independent random sample of 800 Gold Card
members revealed a sample mean $523. Assume population
standard deviation of Preferred Visa as $212 and of Gold Card
members as $185. (Holders of both the Gold Card and Preferred
Visa were excluded from the study.) Is there evidence to
conclude that the average monthly charge in the entire
population of American Express Gold Card members is different
8-26
Example
Population 1: Preferred
Visa H : 1  2 
0
0H :  
n =
1 
1 1 2
1200 0
x =
1
( x1  x 2)  (  1   2)0
452= z   (452  523) 
2
1
0
212
2 2122  1852
n n 1200 800
11  2
Population 2 : Gold 2
Card  71
  71
80.2346 8.96 7.926

n =
2
800
x =
p - value : p(z < -7.926) 
523
 =
2
2
0
185
H is rejected at any common level of significance
0
8-27
Example
Since the value of the test statistic is
far below the lower critical point, the
0.4
null hypothesis may be rejected, and
0.3
we may conclude that there is a
0.2
statistically significant difference
f(z
between the average monthly charges

)
0.1
0.0
of Gold Card and Preferred Visa
z
-z0.01=-2.576
z0.01=2.576
0
cardholders.
Rejection Nonrejection Rejection
Test Statistic=-7.926
Example
Suppose that the makers of Duracell batteries want to
demonstrate that their size AA battery lasts an average of at
least 45 minutes longer than Duracell’s main competitor, the
Energizer. Two independent random samples of 100 batteries of
each kind are selected, and the batteries are run continuously
until they are no longer operational. The sample average life for
Duracell is found to be 308 minutes. The result for the
Energizer batteries is 254 minutes. Assume population s.d1 is 84
minutes and population s.d2 67 minutes. Is there evidence to
substantiate Duracell’s claim that its batteries last, on average,
at least 45 minutes longer than Energizer batteries of the same
size?
8-29
Example
Population 1:
Duracell H : 1  2  45
0
n =
1
H
1
: 
1
 
2
 45
100
x =
308
 =
1 ( x  x )  (   ) (308  254)  45
 
1
z  1 2 1 2 0 
84 2 2
  842  672
1  2
100 100
Population 2 : n
1
n
Energizer 
9
 2
9
 0.838
115 .4 5 10.75
n =
2
100
x = p - value : p(z > 0.838) = 0.201
254
2
H m a y not be rejected at any c o m m o n
 = 67
2
0
level of significance
8-30
A Test for the Difference between Two Population

Means: Assuming Equal Population Variances
◼ If we might assume that the population variances 12 and 22 are equal (even though
unknown), then the two sample variances, s12 and s22, provide two separate estimators of
the common population variance. Combining the two separate estimates into a pooled
estimate should give us a better estimate than either sample variance by itself.
Deviation from the Deviation from the

mean. One for each mean. One for each
sample data point. sample data point.
}
}
* * * * * ** * * * ** * * *** * *
* * ** * * Sample 2
x2
*
x
From sample 1 we get1the estimate s1 with
2 From sample 2 we get the estimate s22 with
(n -1) degrees of freedom.
Sample 1 (n2-1) degrees of freedom.
1
From both samples together we get a pooled estimate, sp2 , with (n1-1) + (n2-1) = (n1+ n2 -2)
total degrees of freedom.
8-31
Pooled Estimate of the Population

Variance
A pooled estimate of the common population variance, based on a sample

variance s12 from a sample of size n1 and a sample variance s22 from a sample
of size n2 is given by:
(n  1)s 2  (n 
s 2p 
1)s
1 12
n1  n2 
2 2
2
The degrees of freedom associated with this estimator is:
df = (n1+ n2-2)
The pooled estimate of the variance is a weighted average of the two

individual sample variances, with weights proportional to the sizes of the
two samples. That is, larger weight is given to the variance from the larger
sample.
8-32
Using the Pooled Estimate of the

Population Variance
2 1 1 
The estimate of the standard deviation of (x 1  x 2 ) is given by: sp  
 1n n 2
Test statistic for the difference between two population means, assuming equal
population variances:
(x1  x2 )  ( 1   2 ) 0
t=
 1  1
s2p  
 1 2
n n
where ( 1   2 ) 0 is the difference

 between the two population means under the null
hypothesis (zero or some other number D).
The number of degrees of freedom of the test statistic is df = (n1  n2  2) (the

2
number of degrees of freedom associated with sp , the pooled estimate of the
population variance.
Example
Changes in the price of oil have long been known to affect the economy
of the United States. An economist wants to check whether the price of a
barrel of crude oil affects the consumer price index (CPI), a measure of
price levels and inflation. The economist collects two sets of data: one
set comprises 14 monthly observations on increases in the CPI, in
percentage per month, when the price of crude oil is $66.00 per barrel;
the other set consists of 9 monthly observations on percentage increase
in the CPI when the price of crude oil is $58.00 per barrel. The economist
assumes that her data are a random set of observations from a
population of monthly CPI percentage increases when oil sells for $66.00
per barrel, and an independent set of random observations from a
population of monthly CPI percentage increases when oil sells for $58.00
per barrel. She also assumes that the two populations of CPI percentage
increases are normally distributed and that the variances of the two
populations are equal. Considering the nature of the economic variables
in question, these are reasonable assumptions. If we call the population
of monthly CPI percentage increases when oil sells for $66.00 population
1, and that of oil at $58.00 per barrel population 2, then the economist’s
data are as follows: sample mean1 = 0.317%, s1= 0.12%, n1 = 14;
sample mean
= 0.210%, s2 = 0.11%, n2 = 9. Our economist is faced with the question:
8-34
Solution
Population1: Oil price =
H 0:  1   2 
$66.00 n1 =14
0 H1: 1   ( x21  x2 )  ( 1   2 )
x1 = 0.317% t

0
(n1 0 1)s12  (n2  1)s22    1 
s1 = 0.12% 1 n1  n2  2

  n1 n2
Population 2: Oil price =
0.10 0.10
$58.00 n = 9  7  7 
0.049 2.154 
2
x2 = 0.0024
7 7
s0.21%
2 = 0.11%
Critical point: t =
2.080 0.02
df = (n1 n2  2)  (14 9  2) 5 at the 5% level of significance
H 0 may be rejected
 21
Example
The manufacturers of compact disk players want to test whether a
small price reduction is enough to increase sales of their product.
Randomly chosen data on 15 weekly sales totals at outlets in a
given area before the price reduction show a sample mean of
$6,598 and a sample standard deviation of $844. A random
sample of 12 weekly sales totals after the small price reduction
gives a sample mean of $6,870 and a sample standard deviation
of $669. Is there evidence that the small price reduction is
enough to increase sales of compact disk players? Assume that
population is normally distributed and population variances are
assumed normal.
8-36
Example 8-6
H :  1  0
Population 1: Before 0 2
H :
Reduction  0
n1 = 15 1 2
1
x1 = ( x  x1)  (  2  1 )0
t 2
$6598  (n  1)s 2  (n  1)s 2   1 1 

s1 =  1
 1 2 2   
n n 2  n n 
$844  1 2
 1 2
(6870  6598)  0
Population 2: After Reduction 
n = 12  (14)844 2  (11)669 2   1  1 
 
2x = $6870   15  12 
2   15 12 
2
s2 = 272 272
  0.91
$669  89375.25 298.96
df = (n  n  2)  (15  12  2)  25
1 2
Critical point : t = 1.316
0.10
H may not be rejected even at the10% level of significance

0
8-37
Confidence Intervals for the Difference

between Two Population Means
A large-sample (1-)100% confidence interval for the difference

between two population means, 1- 2 , using independent random
samples:
2 2
(x x )z 1 2
1 2
 n
n
1
2
2
8-38
Confidence Intervals Using the Pooled

Variance
A (1-) 100% confidence interval for the difference between two

population means, 1- 2 , using independent random samples and
assuming equal population variances:
2 1  1 
( x1  x2 )  t  sp 
 n1 n2 
2
8-39
A Large-Sample Test for the Difference between

Two Population Proportions
• Hypothesized difference is zero
 I: Difference between two population proportions is 0
• p 1= p 2
» H0: p1 -p2 = 0
» H1: p1 -p2  0
 II: Difference between two population proportions is less than 0
• p 1 p 2
» H0: p1 -p2  0
» H1: p1 -p2 > 0
• Hypothesized difference is other than zero:
 III: Difference between two population proportions is less than D
• p1  p2+D
» H0:p-p2  D
» H1: p1 -p2 > D
8-40
8-4 A Large-Sample Test for the Difference

between Two Population Proportions
• Hypothesized difference is zero

 IV: Difference between two population proportions is greater than
0
• p 1 p 2
» H0: p1 -p2  0
» H1: p1 -p2 < 0
• Hypothesized difference is other than zero:
 V: Difference between two population proportions is greater than
D
• p1  p2+D
» H0:p-p2  D
» H1: p1 -p2 < D
Two Population Proportions, p1-p2
Statistic is computed by
Two Population Proportions, p1- p2
Modified formula for testing the hypotheses is,
8-43
Comparisons of Two Population Proportions When the

Hypothesized Difference Is Zero: Example
Finance incentives by the major automakers are reducing banks’

share of the market for automobile loans. Suppose that in 2000,
banks wrote about 53% of all car loans, and in 2007, the banks’
share was only 43%. Suppose that these data are based on a
random sample of 100 car loans in 2000, where 53 of the loans
were found to be bank loans; and the 2007 data are also based
on a random sample of 100 loans, 43 of which were found to be
bank loans. Carry out a two-tailed test of the equality of banks’
share of the car loan market in 2000 and in 2007 at α = 0.10
8-44

Hypothesized Difference Is Zero: Example
Population 198
1: 0
H 0 : p1  p 2  0
xn1 == 100 H 1 : p1  p 2  0 0.53  0.43

53 z 
p1 =
p1  p
p( (1  1  1  (.48)(.52)  1 1
 0.53   
Population 199 ) p 2 )  0 n1 n2  100
 0.10 100
2:
n = 5  
2 1.415
100
x2 = 43
0.10 0.07065

Critical point: z
0.004992 = 1.645
p
2
= 0.05
0.43 53 
x1 + x2 H 0 may not be rejected even at a 10%
p  43 
 n n 100  0.48 level of significance.
1 2 100
8-45
Carrying Out the Test
Standard Normal Distribution Since the value of the test

statistic is within the
0.4
nonrejection region, even at

0.3
0.2
a 10% level of significance,
f(z
)
0.1
0.0
we may conclude that there
z
0
is no statistically significant
-z
0.05
Rejection
=-1.645
Nonrejection Rejection
z =1.645
0.05
difference between banks’
Test Statistic=1.415 shares of car loans in 2000
and 2007.
8-46

Hypothesized Difference Is Not Zero: Example
From time to time, BankAmerica Corporation comes out with its
Free and Easy Travelers Cheques Sweepstakes, designed to
increase the amounts of BankAmerica traveler’s checks sold.
Since the amount bought per customer determines the
customer’s chances of winning a prize, a manager hypothesizes
that, during sweepstakes time, the proportion of BankAmerica
traveler’s check buyers who buy more than $2,500 worth of
checks will be at least 10% higher than the proportion of
traveler’s check buyers who buy more than $2,500 worth of
checks when there are no sweepstakes. A random sample of 300
traveler’s check buyers, taken when the sweepstakes are on,
reveals that 120 of these people bought checks for more than
$2,500. A random sample of 700 traveler’s check buyers, taken
when no sweepstakes prizes are offered, reveals that 140 of
these people bought checks for more than $2,500. Conduct the
hypothesis test at α = 0.001
8-47

Hypothesized Difference Is Not Zero: Example 8-8
Population 1: With
H 0 : p1  p2  0.10
Sweepstakes
n1 = 300
H1: p1  p2  0.10 p )
( p
x1 = 120 z 1 2
D p (1  p
p
1 =  p(1
1  1 2 2
0.40 p )  ) 
 n1 n2 
(0.40  0.20)  0.1
Population 2: No  0.10  0  3.118
Sweepstakes
n = 700 0.03207
2  (0.40)(0.60)  (0.20)(.80)
x2 = 140  300 700

p 2 =  point: z
Critical =
3.09 0.001
H 0 may be rejected at any
0.20
common level of
8-48
Carrying Out the Test

Since the value of the test statistic is
above the critical point, even for a
0.4
level of significance as small as
0.3
0.001, the null hypothesis may be
0.2
rejected, and we may conclude that
the proportion of customers buying at
f(z
0.1
0.0
least
$2500 of travelers checks is at least
z
0
z0.001=3.09
Nonrejection
Region
Rejection 10% higher when sweepstakes are
Region
Test Statistic=3.118 on.
8-49
Confidence Intervals for the Difference

between Two Population Proportions
A (1-) 100% large-sample confidence interval for the difference

between two population proportions:
 p1 1 2 2 (1 
( p1  p 2 )  
p

 )
z
 n1 n2  p
2 (1 
p )
8-50
The F Distribution and a Test for

Equality of Two Population Variances
The F distribution is the distribution of the ratio of two chi-

square random variables that are independent of each other,
each of which is divided by its own degrees of freedom.
An F random variable with k1 and k2 degrees of freedom:
k
Fk1, k 2 
2
11
22
k2
8-51
The F Distribution
• The F random variable
cannot be negative, so it F Distributions with different Degrees of
Freedom
is bound by zero on the
left.
1.0 F(25,30)
f(F)
F(10,15)
• The F distribution is skewed
0.5
to the right.
• The F distribution is 0.0 F(5,6)
identified the number of 0 1 2 3 4 5

F
degrees of freedom in the

numerator, k1, and the
number of degrees of
freedom in the
denominator, k2.
The χ2 Distribution
The F Distribution
F (k1, k2, 1-α/2) F (k1, k2, α/2)
= 1
F (k1 , k2, 1-α/2)
F (k2, k1, α/2)
The left-hand critical point to go along with F(k1,k2) is given by: 1

F k 2,k1 
Where F(k1,k2) is the right-hand critical point for an F random variable with the
reverse number of degrees of freedom.
8-54
Using the Table of the F Distribution :

F(7, 11),  = 0. 05
Critical Points of the F Distribution Cutting Off a F Distribution with 7 and 11 Degrees of
Right-Tail Area of 0.05
Freedom
k1 1 2 3 4 5 6 7 8 9 0.7
k2 0.6
1 161.4 199.5 215.7 224.6 230.2 234.0 236.8 238.9 240.5 0.5
2 18.51 19.00 19.16 19.25 19.30 19.33 19.35 19.37 19.38
f(F)
3 10.13 9.55 9.28 9.12 9.01 8.94 8.89 8.85 8.81 0.4
4 7.71 6.94 6.59 6.39 6.26 6.16 6.09 6.04 6.00
0.3
5 6.61 5.79 5.41 5.19 5.05 4.95 4.88 4.82 4.77
6 5.99 5.14 4.76 4.53 4.39 4.28 4.21 4.15 4.10 0.2
7 5.59 4.74 4.35 4.12 3.97 3.87 3.79 3.73 3.68
0.1
8 5.32 4.46 4.07 3.84 3.69 3.58 3.50 3.44 3.39
9 5.12 4.26 3.86 3.63 3.48 3.37 3.29 3.23 3.18 0.0 F
10 4.96 4.10 3.71 3.48 3.33 3.22 3.14 3.07 3.02 0 1 2 3 4 5
11 4.84 3.98 3.59 3.36 3.20 3.09 33..001 2.95 2.90
1
12 4.75 3.89 3.49 3.26 3.11 3.00 2.91 2.85 2.80 F0.05=3.01
13 4.67 3.81 3.41 3.18 3.03 2.92 2.83 2.77 2.71
14 4.60 3.74 3.34 3.11 2.96 2.85 2.76 2.70 2.65
15 4.54 3.68 3.29 3.06 2.90 2.79 2.71 2.64 2.59
8-55
Critical Points of the F Distribution:

F(6, 9),  = 0.10
F Distribution with 6 and 9 Degrees of The right-hand critical point

Freedom
0.7
read directly from the table of
0.6 0.90 the F distribution is:
0.05
0.5
F(6,9) =3.37
f(F)
0.4
0.3
0.05
0.2
0.1 The corresponding left-hand critical

0.0
0 1 2 3 4 5 F
point is given by:
1 1
F0.95=(1/4.10)=0.2439 F0.05=3.37
  0.2439
4.10
F9,6
8-56
Test Statistic for the Equality of Two

Population Variances
Test statistic for the equality of the variances of two normally
distributed populations:
2
F  s1
n1  1 , n 2 1
s22
 I: Two-Tailed Test
• 1 = 2
• H0: 1 = 2
• H1:   2
 II: One-Tailed Test
• 12
• H0: 1  2
• H1: 1  2
Example
One of the problems that insider trading supposedly causes is
unnaturally high stock price volatility. When insiders rush to buy a
stock they believe will increase in price, the buying pressure
causes the stock price to rise faster than under usual conditions.
Then, when insiders dump their holdings to realize quick gains,
the stock price dips fast. Price volatility can be measured as the
variance of prices. An economist wants to study the effect of the
insider trading scandal and ensuing legislation on the volatility of
the price of a certain stock. The economist collects price data for
the stock during the period before the event (interception and
prosecution of insider traders) and after the event. The economist
makes the assumptions that prices are approximately normally
distributed and that the two price data sets may be considered
independent random samples from the populations of prices
before and after the event. As we mentioned earlier, the theory of
stock.
financeThe 25 daily
supports thestock pricesassumption.
normality before the 1event
(The give s 2 9.3 of
assumption
the 24 stock
random
(dollars pricesmay
sampling
squared), after
and bethe event
somewhat
2 give s 2 3.0 (dollars
problematic squared).
in this case, but
later
Conduct
α = we the
will test
dealat
with time-dependent observations more
0.05.
effectively.) Suppose that the economist wants to test whether
8-58
Solution
Population 1 : Before
n = 25
1
s 12  9 . 3 2 2
2
H : 
Population 2 : After 0 1 21
n = 24
2 2 2
H 1:  1   2
s 22  3 . 0
2
1s 9.3
F F    3.1
  0.05 n1  1, n2  1 24,23 s2 3.0
2
F  2.01
24,23
  0.01 H 0 may be rejected at a 1% level of significance.
F  2.70
24,23
8-59
Solution
Distribution with 24 and 23 Degrees of Freedom

Since the value of the test
statistic is above the critical
0.7
0.6
0.5 point, even for a level of

significance as small as 0.01,
f(F)
0.4
0.3
0.2 the null hypothesis may be

0.1
0.0 F
rejected, and we may conclude
0 1 2 3 4 5
that the variance of stock
F0.01=2.7
Statistic=3.1
Test
prices is reduced after the
interception and prosecution
of inside traders.
Example
Suppose a machine produces metal sheets that are specified to be 22 millimeters
thick. Because of the machine, the operator, the raw material, the manufacturing
environment, and other factors, there is variability in the thickness. Two machines
produce these sheets. Operators are concerned about the consistency of the two
machines. To test consistency, they randomly sample 10 sheets produced by
machine 1 and 12 sheets produced by machine 2. The thickness measurements of
sheets from each machine are given below. Assume sheet thickness is normally
distributed in the population. Test to determine whether the variance from each
sample comes from the same population variance (population variances are
equal) at α = .05
Solution
= 1 = 1 = 0.255 F (0.975,9,11) = 0.255

F (0.975, 9, 11 )
3.91
F (.025, 11, 9)
An examination of the sample variances reveals that the

variance from machine 1 measurements is greater than
that from machine 2 measurements. The operators and
process managers might want to examine machine 1
further; an adjustment may be needed or some other
reason may be causing the seemingly greater variations
on that machine.
Example
According to Runzheimer International, a family of
four in Manhattan with $60,000 annual income
spends more than $22,000 a year on basic goods
and services. In contrast, a family of four in San
Antonio with the same annual income spends only
$15,460 on the same items. Suppose we want to
test to determine whether the variance of money
spent per year on the basics by families across the
United States is greater than the variance of money
spent on the basics by families in Manhattan—that
is, whether the amounts spent by families of four in
Manhattan are more homogeneous than the
amounts spent by such families nationally.
Suppose a random sample of eight Manhattan
families produces the figures in the table, which are
given along with those reported from a random
sample of seven families across the United States.
Complete a hypothesis-testing procedure to
determine whether the variance of values taken
from across the United States can be shown to be
greater than the variance of values obtained from
families in Manhattan. Let = .01. Assume the
amount spent on the basics is normally distributed
in the population.
Solution
The variance for families in the United States is greater than the
variance of families in Manhattan. Families in Manhattan are more
homogeneous in amount spent on basics than families across the
United States.
9-1
Analysis of Variance
9-2
Using Statistics
• ANOVA (ANalysis Of VAriance) is a statistical method for determining
the existence of differences among several population means.
 ANOVA is designed to detect differences among means from
populations subject to different treatments (different populations)
 ANOVA is a joint test
◼ T h e equality of several population means is tested
simultaneously or jointly.
ANOVA tests for the equality of several population means by

looking at two estimators of the population variance (hence,
analysis of variance).
One-Way ANOVA
Suppose a researcher decides to analyze the effects of the four
machine operators on the valve opening measurements of valves
produced in a manufacturing plant. Is there a significant difference
in the mean valve openings of 24 valves produced by the four
operators?
9-4
The Hypothesis Test of Analysis of

Variance
The hypothesis test of analysis of variance:
H0: 1 = 2 = 3 = 4 = ... r
H1: Not all i (i = 1, ..., r) are equal
There are r populations, or treatments, under study.

We draw an independent random sample from each of the r
populations. The size of the sample from population i (i =1, .
. . , r) is ni , and the total sample size is
N = n1 + n2 + …… nr
One-Way ANOVA
Suppose a researcher decides to analyze the effects of the four machine
operators on the valve opening measurements of valves produced in a
manufacturing plant. Is there a significant difference in the mean valve openings
of 24 valves produced by the four operators?
Is it possible to analyze the four samples by using a t test for the difference in two
sample means? These four samples would require 4C2 = 6 individual t tests to
accomplish the analysis of two groups at a time. If α= .05 for a particular test,
there is a 5% chance of committing a Type I error. In this problem, with
six t tests, the error rate compounds, so when the analyst is finished with the
problem there is a much greater than .05 chance of committing a Type I error
9-6

Variance
The test statistic of analysis of
variance:
Estimate of variance based on means from r samples
F(r-1, n-r) = Estimate of variance based on all sample observations
That is, the test statistic in an analysis of variance is based

on the ratio of two estimators of a population variance, and
is therefore based on the F distribution, with (r-1) degrees
of freedom in the numerator and (n-r) degrees of freedom in
the denominator.
 Analysis of variance tests are always one-

tailed tests with the rejection region in the
upper tail. The decision rule is to reject the
null hypothesis if the observed F value is
greater than the critical F value
9-7

Variance (continued): Assumptions
• We assume independent random sampling from each of the r

populations
• We assume that the r populations under study:
 are normally distributed,
 with means i that may or may not be equal,
 but with equal variances,  2.
………
1 2 r
Population 1 Population 2 Population r
9-8
The ANOVA Test Statistic for r = 4 Populations and n = 54 Total
Sample Observations
Suppose we have 4 populations, from each of which we

draw an independent random sample, with n1 + n2 + n3 + n4
= 54. Then our test statistic is:
Estimate of variance based on means from 4 samples
F(4-1, 54-4)= F(3,50) = Estimate of variance based on all 54 sample observations
F Distribution with 3 and 50 Degrees of

Freedom
0.
7
0. The nonrejection region (for =0.05)in this instance is
6
F  2.79, and the rejection region is F > 2.79. If the
f(F
0.
)
5
test statistic is less than
0.
4 =0.05 2.79 we would not reject the null hypothesis, and we
0.
3
would conclude the 4 population means are equal. If
0.
0 1 2 3 4 5 the test statistic is greater than 2.79, we would reject
F(3,50)
2.79
2
the null hypothesis and conclude that the four
0.
1 population means are not equal.
The Hypothesis Test of ANOVA
Variance
The main principle behind the analysis of
variance is,
Variance
The ANOVA principle thus
says
The Sum-of-Squares Principle
The sum-of-squares total (SST) is the sum of the two terms: the sum of squares for
treatment (SSTR) and the sum of squares for error (SSE).
SST = SSTR + SSE
• SST is the total amount of variation within the data set.

• SSTR is that part of the variation within the data that is due to differences among the
groups. It is sometimes called the sum of squares between (variation among the
groups. It is also called the explained variation (because it is the part of the total
variation that can be explained by the fact that the data points belong to several
different groups)
• SSE is that part of the variation within the data that is due to error—the part that
cannot be explained by differences among the groups. It is then called the
unexplained
Degrees of Freedom
The number of degrees of freedom associated with SSE is n - r.
The number of degrees of freedom associated with SST is n - 1.

The number of degrees of freedom associated with SSTR is r - 1.
The Mean Squares

In finding the average squared deviations due to treatment and to error, we divide
each sum of squares by its degrees of freedom. We call the two resulting averages
mean square treatment (MSTR) and mean square error (MSE),
When the null hypothesis of ANOVA is true and all r

population means are equal, MSTR and MSE are two
independent, unbiased estimators of the common
population variance σ2.
Example
Example
ANOVA Table
Since 37.62 is much greater than 8.65 at α = 0.01, we can therefore

Problem
Suppose a researcher decides to analyze the effects of the four
machine operators on the valve opening measurements of valves
produced in a manufacturing plant. Is there a significant difference
in the mean valve openings of 24 valves produced by the four
operators? Test at α = 0.05
Solution
Problem
A company has three manufacturing plants, and company officials
want to determine whether there is a difference in the average age
of workers at the three locations. The following data are the ages of
five randomly selected workers at each plant. Perform a one-way
ANOVA to determine whether there is a significant difference in the
mean ages of the workers at the three plants. Use = .01 and note
that the sample sizes are equal.
Solution
Solution
The decision is to reject the null hypothesis because the observed F

value of 39.80 is greater than the critical table F value of 6.93.
There is a significant difference in the mean ages of workers at the

three plants.

Decision Science

Загружено:

Сведения о документе

Исходное описание:

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

Decision Science

Загружено:

Авторское право:

Доступные форматы

DECISION

and draw meaningful inferences from data that

the running, for example, a department, a

1-1. Using Statistics (Two Categories)

Types of Data - Two Types

 Nationality  Number of points scored on

• Statistical techniques can be separated into two categories:

Samples and Populations

 A population consists of the set of all measurements

Simple Random Sample

 Sampling from the population is often done

random sample or just a random sample.

Samples and Populations

Population (N) Sample (n)

⚫Census of a population may

• Type of measure that is used to describe a set of data yield

◼ The population mean is represented by ( μ). The sample mean is

The magazine Forbes publishes annually a

2. Calculate the percentile location (P ) by (n+1)P /100 where, n is

Billions Sorted Billions

 To find the 80th percentile, determine the data

 To find the 90th percentile, determine the data

Quartiles – Special Percentiles

 Quartiles are the percentage points that

Quartiles and Interquartile Range

 The first quartile, Q1, (25th percentile) is

Example : Finding Quartiles

Sorted (n+1)P/100 Quartiles

It is the 50th percentile.

• The mode is the most frequently occurring value in a set of data.

a. 5.40 1.10 0.42 0.73 0.48 1.10 Mode is 1.10

 Set I : 1,2,3,4,5,6,6,7,8,9,10,11 Mean ?

 Dividing data into groups or classes or intervals

 Table with two columns listing:

 Sum of frequencies is number of observations

Example : Frequency Distribution

0 to less than 100 30 0.163

• Example of relative frequency: 30/184 = 0.163

Cumulative Frequency Distribution

0 to less than 100 30 0.163

The cumulative frequency of each group is the sum of the

Class Interval Frequency

The skewed portion is the long, thin part of the curve

2-1 Probability is:

 A quantitative measure of uncertainty

Types of Probability - Classical

Types of Probability - Subjective

 The subjective method of assigning probability is based on the

 It is a process that produces outcomes.

 It is an outcome of an experiment, the experiment

Basic Definitions (Continued)

Sets: A Intersecting with B

Basic Definitions (Continued)

• Mutually exclusive or disjoint sets

Mutually Exclusive or Disjoint Sets

Sets have nothing in common

Basic Definitions (Continued)

2-3 Basic Rules for

Basic Rules for Probability

The size of the population, N, is 6, the six sides of the

S represent the event “more storage/ filing space.”

P (N U S) = P (N) + P (S) = .70 + .67 = 1.37