Вы находитесь на странице: 1из 30

Business Statistics: Communicating with Numbers

By Sanjiv Jaggia and Alison Kelly

McGraw-Hill/Irwin

Copyright 2013 by The McGraw-Hill Companies, Inc. All rights reserved.


8-1

8.1 Point Estimators and Their


Properties
Point Estimator

A function of the random sample used to make inferences


about the value of an unknown population parameter.
For example, is a point estimator for and is a point
X
estimator for p.
P

Point Estimate

The value of the point estimator derived from a given


sample.
For example,
is a point estimate of the population
mean for all ultra-green
x 96.5 cars.

8-2

8.1 Point Estimators and Their


Properties

Example:

8-3

8.2 Confidence Interval of the


Population
Mean When Is Known

Confidence Intervalprovides a range of


values that, with a certain level of confidence,
contains the population parameter of interest.

Also referred to as an interval estimate.

Construct a confidence interval as:


Point estimate Margin of error (e.g. x D).

Margin of error (D) accounts for the variability of


the estimator and the desired confidence level of
the interval.
8-4

Interval Estimation of a
Population Mean
Margin of error is the maximum difference between
sample and population means D=| x - |
P(-D x +D)=.95
states that 95% of sample means (
interval where D= z/n since

x)

will be in this

=D/(/n)

95% is called confidence coefficient (C)

8-5

Interval Estimation of a
Population Mean
This also means that in 95% of cases will be inside
the interval
D

This interval is called confidence interval (CI)


(e.g., 95% confidence interval)

8-6

Interval Estimation of a
Population Mean (margin of
error D)
C is called confidence coefficient (probability
that the statement is correct)
=1-C is called significance level
(probability that the statement is incorrect
probability of an error)

8-7

Interval Estimation of a Population


Mean (margin of error D)
D=zn
To find z: C/2 +.5 or 1- /2 and look for z in
the z-table.
Confidence interval for the population mean is:
x -D, x + D or

x z 2

8-8

8.2 Confidence Interval of the


Population
Mean
Example:
Constructing
a Confidence Interval
When
Is Known
for When is Known

A sample of 25 cereal boxes of Granola Crunch, a


generic brand of cereal, yields a mean weight of
1.02 pounds of cereal per box.
Construct a 95% confidence interval of the mean
weight of all cereal boxes.
Assume that the weight is normally distributed
with a population standard deviation of 0.03
pounds.
8-9

8.2 Confidence Interval of the


Population
Example (continue)
Mean When Is Known
LO
8.3

This is what we know:

n 25, x 1.02 pounds

= 1 .95 .05, z 2 1.96


0.03

Substituting these values, we get


x 1.96

n 1.02 1.96 0.03

25 1.02 0.012

or, with 95% confidence, the population mean


weight of all cereal boxes falls between 1.008 and
1.032 pounds.
8-10

8.2 Confidence Interval of the


Population
Mean
Interpreting
WhenaConfidence
Is KnownInterval
LO
8.3

Interpreting a confidence interval requires care.


Incorrect: The probability that falls in the
interval is 0.95.

Correct: If numerous samples of size n are drawn


from a given population, then 95% of the intervals
formed by the formula x z 2 n will contain .
Since there are many possible samples, we
will be right 95% of the time, thus giving us
95% confidence.
8-11

8.2 Confidence Interval of the


Population
Mean When Is Known
The Width of a Confidence Interval is equal
to two margins of error 2D = 2 z n
2

The width of the confidence interval is


influenced by the:
Standard deviation
Sample size n
confidence level 100(1 )%.

8-12

8.2 Confidence Interval of the


Population
When
Is Known
Mean
Width of
a Confidence
Interval is influenced by :
LO
8.3

I. For a given confidence level 100(1 )% and sample size


n, the width of the interval is wider, the greater the
population standard deviation .

Example: Let the standard deviation of the population of


cereal boxes of Granola Crunch be 0.05 instead of 0.03.
Compute a 95% confidence interval based on the same
sample information.
x z 2

n 1.02 1.96 0.05

25 1.02 0.20

This confidence interval width has increased from 0.024 to


2(0.020) = 0.040.
8-13

8.2 Confidence Interval of the


Population
When
Is Known
Mean
Width of
a Confidence
Interval is influenced by n:
LO
8.3

II. For a given confidence level 100(1 )% and population


standard deviation , the width of the interval is wider, the
smaller the sample size n.

Example: Instead of 25 observations, let the sample be


based on 16 cereal boxes of Granola Crunch. Compute a
95% confidence interval using a sample mean of 1.02
pounds and a population standard deviation of 0.03.
x z 2

n 1.02 1.96 0.03

16 1.02 0.015

This confidence interval width has increased from 0.024 to


2(0.015) = 0.030.
8-14

8.2 Confidence Interval of the


Population
When
Is Known
Mean
Width of
a Confidence
Interval is influenced by C:
III. For a given sample size n and population standard
deviation , the width of the interval is wider, the greater
the confidence level C=100(1 )%.

Example: Instead of a 95% confidence interval, compute a


99% confidence interval based on the information from the
sample of Granola Crunch cereal boxes.
x z 2

n 1.02 2.575 0.03

25 1.02 0.015

This confidence interval width has increased from 0.024 to


2(0.015) = 0.030.

8-15

8.2 Confidence Interval of the


Population
When Is Known
Mean
Example:

8-16

8.3 Confidence Interval of the


Population
Mean When Is Unknown

If we do not know population standard deviation ,


we use sample standard deviation s and t-table
T

X
S

8-17

8.3 Confidence Interval of the


Population
Mean
When is Is
Unknown
The t distribution
a normal
distribution with
mean equal to zero but its standard deviation
depends on degrees of freedom (n-1).
As the size of the sample increases, the
difference between the t distribution and z
distribution becomes smaller and smaller.

8-18

8.3 Confidence Interval of the


Population

The tdfWhen
Distribution
Various Degrees of
Mean
Iswith
Unknown
Freedom

8-19

8.3 Confidence Interval of the


Population
Example: Compute t
for = 0.025 using 2, 5,
Mean When Is,dfUnknown
and 50 degrees of freedom.
Solution: Turning to the Students t Distribution
table in Appendix A, we find that
For df = 2, t
0.025,2 = 4.303.

For df = 5, t0.025,5 = 2.571.

For df = 50, t0.025,50 = 2.009.


Note that the tdf values change with the degrees
of freedom. Further, as df increases, the tdf
distribution begins to resemble the z distribution.

8-20

8.3 Confidence Interval of the


Population
Mean When Is Unknown

Constructing a Confidence Interval for


When is Unknown is computed as:

x t 2,df s

or equivalently,

x t 2,df s

n , x t 2,df s

where s is the sample standard deviation.

8-21

8.3 Confidence Interval of the


Population

Example:When
Recall that
Beane wants to estimate the
Mean
Jared
Is Unknown

mean mpg of all ultra-green cars. Use the sample of 25 cars


to construct a 90% confidence interval of the population
mean. Assume that mpg follows a normal distribution.

Solution: Since the population standard deviation is


unknown, the sample standard deviation s was
calculated from the sample as 10.70.
t.05.24 =1.711 for /2 =(1-.90)2=.05 and degrees of freedom
d.f.=n-1=25-1=24
As a result, the 90% confidence interval is

x t 2,df s

n 96.52 1.711 10.70

25 96.52 3.66
8-22

8.5 Selecting a Useful Sample Size

Precision in interval estimates is implied by


a low margin of error.
The larger n reduces the margin of error for
the interval estimates.
How large should the sample size be for a
given margin of error?

8-23

8.5 Selecting a Useful Sample


Size
Selecting n to Estimate

Consider a confidence interval for with a


known and let D denote the desired margin of
error.
Since D z 2
n
z 2
we may rearrange to get n

Use is a reasonable estimate of in the planning


stage if it is unknown

8-24

8.5 Selecting a Useful Sample


Size
Example: Recall that Jared Beane wants to

construct a 90% confidence interval of the mean


mpg of all ultra-green cars.

Suppose Jared would like to constrain the margin of error


to within 2 mpg.
How large a sample does Jared need to compute the 90%
confidence interval of the population mean using standard
deviation from previous sample of 10.50 as planning stage
estimate?
2

2
z 2
1.645 10.50
n
74.58 or 75

8-25

8.4 Confidence Interval of the


Population
Proportion

Let the parameter p represent the proportion


of successes in the population, where
success is defined by a particular outcome.
P is the point estimator of the population
proportion p.
By the central limit theorem, P can be
approximated by a normal distribution for
large samples (i.e., np > 5 and n(1 p) > 5).
8-26

8.4 Confidence Interval of the


Population

Thus, a 100(1)% confidence interval of


Proportion
the population proportion is
p z 2

p 1 p
n

or

p 1 p

p z 2

, p z 2

p 1 p

where p is the sample proportion and


is the standard deviation of the proportion

8-27

8.4 Confidence Interval of the


Population
Example: Recall that Jared Beane wants to estimate the
Proportion

proportion of all ultra-green cars that obtain over 100 mpg.


Use the sample information to construct a 90% confidence
interval of the population proportion.
Solution: Note that in addition, the normality assumption
is met since np > 5 and n(1 p) > 5. Sample proportion is
p 7 25 0.28., n=25,
.05,24=1.645. Confidence interval is
defined as:

8-28

8.5 Selecting a Useful Sample


Size
Selecting n to Estimate p

Consider a confidence interval for p and let D


denote the desired margin of error.
Since
where p is the
p 1 p
sample proportion
Dz
2

z 2
we may rearrange to get n
p 1 p
D

Since p comes from a sample, we must use a


reasonable estimate of p (e.g. from
p previous
studies).
8-29

8.5 Selecting a Useful Sample


Size
Example: Recall that Jared Beane wants to

construct a 90% confidence interval of the


proportion of all ultra-green cars that obtain over
100 mpg.

Jared does not want the margin of error to be more than


0.10.

How large a sample does Jared need for his analysis of


the population proportion if he uses population proportion
estimate of .50?
2

z 2
1.645

n
0.50 1 0.50 67.65 or 68
p 1 p

0.10
D
8-30

Вам также может понравиться