6 More Inference about a Population

Inference about a Normal Population Mean: Variance Unknown

Section 6.1 Inference about a Normal Population Mean: Variance

Unknown

2
Assumption The population follows N( , ).

Let X be the sample mean of a sample taken from

the population, S be the sample standard deviation
𝑋−𝜇
and n be the sample size. It can be proved that
𝑆/√𝑛
follows the t distribution with
n 1 degrees of freedom.

This graph shows the probability density

functions of N(0, 1) and some t(n), the
t distribution with n degrees of freedom
where n is a positive integer.

Interval estimation
The 100(1 )% confidence interval for the population mean is
s s
x t / 2, n 1 ,x t / 2, n 1
n n
where the value t , n-1 can be obtained from the t distribution table. Comparing
with the case with known , here we use s instead of and t 2,n 1 instead of z 2.

Density function of the t distribution with n degrees of freedom

This study source was downloaded by 100000787038097 from CourseHero.com on 01-09-2022 20:55:43 GMT -06:00
ch6 1
https://www.coursehero.com/file/113801646/Lecture-Notes-6-v118pdf/
Example 1 A paint manufacturer wants to determine the average drying time of

mean drying time of _

o 66.3 minutes and a standard deviation of 8.4 minutes,
n_n
a new brand of interior wall paint. If for 12 test areas of equal size he obtained a

_
construct a 95% confidence interval for the true population mean assuming
normality.

[Solution] n 12, 𝑥 66.3, s 8.4, 1 0.95 0.05 and

t 2,n 1 t0.025,11 2.201.

The 95% confidence interval for is

8.4 8.4 ,
66.3 2.201 , 66.3 2.201
12 12
that is, (61.0, 71.6).

Hypothesis test
When the null hypothesis is 0, the test statistic is chosen to be

𝑋−𝜇
𝑆/√𝑛

Let be the level of significance.

Test H0 H1 Reject H0 if and only if
Left- 0 𝑋−𝜇0
0 t ,n 1
tailed or 0 𝑆/√𝑛

Right- 0 𝑋−𝜇0
0 t ,n 1
tailed or 0 𝑆/√𝑛

Two- 𝑋−𝜇0
0 0 | | t 2,n 1
tailed 𝑆/√𝑛

For a two-tailed test, the followings are equivalent (for each particular sample):

1. do not reject H0: 0 at the level of significance;

2. the 100(1 )% confidence interval for contains 0.

This study source was downloaded by 100000787038097 from CourseHero.com on 01-09-2022 20:55:43 GMT -06:00
ch6 2
https://www.coursehero.com/file/113801646/Lecture-Notes-6-v118pdf/
 Example 2 A psychologist claims that the mean age at which children start
walking is 12.5 months. Carol wanted to check if this claim is true. She took a
random sample of 18 children and found that the mean age at which these children
started walking was 12.9 months and the standard deviation was 0.80 months.
Using the 1% significance level, can you conclude that the mean age at which all
children start walking is different from 12.5 months?
Ǖstest Nf
[Solution] 。

H0: 12.5, H1: 12.5, 0.01.

The value of the test statistic is

𝑋 − 𝜇0 12.9 − 12.5
= = 2.121
𝑆/√𝑛 0.8/√18

Since |2.121| 2.898 t0.005,17, we do not reject H0. Thus we cannot conclude that
the mean age at which all children start walking is different from 12.5 months.

Example 3 According to the last census in a city, the mean family annual income
was 316 thousand dollars. A random sample of 900 families taken this year
produced a mean family annual income of 313 thousand dollars and a standard
deviation of 70 thousand dollars. At the 2.5% significance level, can we conclude
that the mean family annual income has declined since the last census?

[Solution]
H0: = 316, H1: 316, 0.025.

The value of the test statistic is

𝑋−𝜇0 313−316
= 1.286
𝑆/√𝑛 70/√900

By Excel, t0.025,899 1.963. Since 1.286 1.963 t0.025,899, we do not reject H0.
Thus we cannot conclude that the mean family annual income has declined since
the last census.

This study source was downloaded by 100000787038097 from CourseHero.com on 01-09-2022 20:55:43 GMT -06:00
ch6 3
https://www.coursehero.com/file/113801646/Lecture-Notes-6-v118pdf/
 Example 4 A federal agency responsible for enforcing laws governing weights
and measures routinely inspects packages to determine whether the weight of the
contents is at least as great as that advertised on the package. A random sample of
18 containers whose packaging states that the contents weigh 800 grams was
drawn. The contents were weighed and the results listed below:

780, 791, 793, 799, 794, 775,

797, 795, 779, 806, 782, 789,
792, 787, 792, 798, 805, 791.

Can we conclude that on average the containers are mislabelled? Use 0.1.

[Solution]
H0: = 800, H1: 800, 0.1.

x 791.3889, s 8.4724.

𝑋−𝜇0 791.3889−800
𝑆 = 8.4724 4.31 1.333 - t0.1, 17
√𝑛 √18

Reject H0. We can conclude that on average the containers are mislabelled.

This study source was downloaded by 100000787038097 from CourseHero.com on 01-09-2022 20:55:43 GMT -06:00
ch6 4
https://www.coursehero.com/file/113801646/Lecture-Notes-6-v118pdf/
 Section 6.2 Inference about a Population Proportion Based on Large
Samples

The population proportion, denoted by p, is obtained by taking theoratio of the

number of elements in a population with a specific characteristic to the total
number of elements in the population. The sample proportion, denoted by 𝑝̂ ,
gives a similar ratio of a sample.

Example 1 A total number of 789,654 families live in a city and 563,282 of them
own homes. Then the population proportion of homeowners is

p 563,282 789,654 0.7133.

Suppose a sample of 240 families is taken from this city and 158 of them are
homeowners. Then the sample proportion is

𝑝̂ 158 240 0.6583.

When population proportion is considered, a sample of size n is said to be large if

n p and n (1 p) are both greater than 5. If p is unknown, then n𝑝̂ and n (1 𝑝̂ )
should be both greater than 5 for a sample to be large.

Interval Estimation
Assume that we have a large sample. By the central limit theorem, the 100(1
)% confidence interval for the population proportion is approximately

p (1 p ) p (1 p) .
pˆ z /2 , pˆ z /2
n n

Since p is unknown, p is replaced by 𝑝̂ . Therefore the 100(1 )% confidence

interval for the population proportion is approximately

pˆ (1 pˆ ) pˆ (1 pˆ ) .
pˆ z /2 , pˆ z /2
n n

This study source was downloaded by 100000787038097 from CourseHero.com on 01-09-2022 20:55:43 GMT -06:00
ch6 5
https://www.coursehero.com/file/113801646/Lecture-Notes-6-v118pdf/
 Example 2 In a random sample, 136 of 400 persons given a flu vaccine
experienced some discomfort. Construct a 95% confidence interval for the true
population proportion of persons who will experience some discomfort from the
vaccine.

[Solution] n 400, 𝑝̂ 136 400 0.34. Therefore n𝑝̂ 136 5 and n (1 𝑝̂ )

400 136 5, that is, the sample is large.

0.34(1−0.34)
0.34 1.960√ 0.2936,
400
0.34(1−0.34)
0.34 1.960 √ 0.3864.
400

Therefore the 95% confidence interval is (0.2936, 0.3864).

Hypothesis test
Assume that we have a large sample. The value of the test statistic for H0: p p0
𝑝̂−𝑝0
is . Let be the level of significance. (There are z and z 2 in
√𝑝0 (1−𝑝0 )/𝑛

the following table because the central limit theorem is applied.)

Test H0 H1 Reject H0 if and only if

Left- p p0 𝑝̂−𝑝0
z
p p0
tailed or p p0 √𝑝0 (1−𝑝0 )/𝑛

Right- p p0 𝑝̂−𝑝0
z
p p0
tailed or p p0 √𝑝0 (1−𝑝0 )/𝑛

Two- |𝑝̂−𝑝0 |
z
p p0 p p0 2
tailed √𝑝0 (1−𝑝0 )/𝑛

This study source was downloaded by 100000787038097 from CourseHero.com on 01-09-2022 20:55:43 GMT -06:00
ch6 6
https://www.coursehero.com/file/113801646/Lecture-Notes-6-v118pdf/
 Example 3 When working properly, a machine that is used to make chips for
calculators does not produce more than 4% defective chips. Whenever the machine
produces more than 4% defective chips, it needs an adjustment. A random sample
of 200 chips taken recently from the production line contained 14 defective chips.
Test at the 5% significance level whether or not the machine needs an adjustment.

[Solution]
H0: p = 0.04, H1: p 0.04, 0.05.

n p0 200 (0.04) 8 5,
n (1 p0) n n p0 200 8 5.

Therefore the sample is large.

𝑝̂ 14 200 0.07,
pˆ p0 0.07 0.04
2.1651.
p0 (1 p0 ) / n 0.04(1 0.04) / 200

Since 2.1651 1.645 z0.05, we reject H0 and accept H1. We can conclude that the
machine needs an adjustment.

This study source was downloaded by 100000787038097 from CourseHero.com on 01-09-2022 20:55:43 GMT -06:00
ch6 7
https://www.coursehero.com/file/113801646/Lecture-Notes-6-v118pdf/
Powered by TCPDF (www.tcpdf.org)