Chapter 7. Statistical Intervals For A Single Sample

Chapter 7: Statistical Intervals for a Single
Sample
Course Name: PROBABILITY & STATISTICS

Lecturer: Duong Thi Hong
Hanoi, 2022
1 / 91 Chapter 7: Statistical Intervals for a Single Sample

Content
1 Confidence Interval on the Mean of a Normal Distribution,

Variance Known
Variance Unknown
3 Large-Sample Confidence Interval for a Population
Proportion

Introduction
In the previous chapter we illustrated how a parameter can be estimated

from sample data. However, it is important to understand how good is the
estimate obtained.
Bounds that represent an interval of plausible values for a parameter are
an example of an interval estimate.

Content

Variance Known
Variance Unknown
Proportion

1.1 Development of the Confidence Interval and
Its Basic Properties
Suppose that X1, X2, ..., Xn is a random sample from a normal distribution
with unknown mean µ and known variance σ2 . Then, the sample mean X̄
is normally distributed with mean µ and variance σ2/n. Therefore,
X̄ − µ
Z= √
σ/ n
is a standard normal distribution.

Confidence Interval
A confidence interval (CI) estimate for µ is an interval of the form l ≤

µ ≤ u, where the endpoints l and u are computed from the sample data.
Because different samples will produce different values of l and u, these
end-points are values of random variables L and U , respectively. Suppose
that we can determine values of L and U such that the following probability
statement is true:
P (L ≤ µ ≤ U ) = 1 − α, for 0 ≤ α ≤ 1
The endpoints or bounds l and u are called lower- and

upper-confidence limits, respectively.
1 − α is called the confidence coefficient.
Since Z = σ/X̄−µ
√ has a standard normal distribution, we may write
n
X̄ − µ
P (−zα/2 ≤ √ ≤ zα/2 ) = 1 − α
σ/ n
This results in
σ σ
P (X̄ − zα/2 √ ≤ µ ≤ X̄ + zα/2 √ ) = 1 − α
n n

Confidence Interval on the Mean, Variance Known
If x̄ is the sample mean of a random sample of size n from a normal

population with known variance σ 2 , a 100(1 − α)% CI on µ is given
by
σ σ
x̄ − zα/2 √ ≤ µ ≤ x̄ + zα/2 √
n n
where zα/2 is the upper 100α/2 percentage point of the standard
normal distribution.

Confidence Interval on the Mean, Variance
Known
The situation is illustrated in following figure, which shows several

100(1 − α)% confidence intervals for the mean µ of a normal dis-
tribution. The dots at the center of the intervals indicate the point
estimate of µ (that is, x̄). Notice that one of the intervals fails to
contain the true value of µ. If this were a 95% confidence interval, in
the long run only 5% of the intervals would fail to contain µ.

Known
Example 1
In a sample of 36 randomly selected women, it was found that their

mean height was 65.3 inches. From previous studies, it is assumed
that the standard deviation σ = 2.5. Construct the 90% confidence
interval for the population mean. Let z0.05 = 1.64.

Known
Example 1
In a sample of 36 randomly selected women, it was found that their

mean height was 65.3 inches. From previous studies, it is assumed
that the standard deviation σ = 2.5. Construct the 90% confidence
interval for the population mean. Let z0.05 = 1.64.
Answer: We known σ = 2.5, x̄ = 65.3, n = 36, α = 0.1. Therefore,

σ 2.5
zα/2 √ = z0.05 √ = 0.683
n 36
Hence, the 90% confidence interval for the population mean is
(65.3 − 0.683, 65.3 + 0.683) = (64.617, 65.983)

Known
Example 2
An economist is interested in studying the incomes of consumers

in a particular region. The population standard deviation is known
to be $1000. A random sample of 64 individuals resulted in an
average income of $21000. What is the width of the 95% confidence
interval?. Let z0.025 = 1.96.

Known
Example 2
An economist is interested in studying the incomes of consumers

in a particular region. The population standard deviation is known
to be $1000. A random sample of 64 individuals resulted in an
average income of $21000. What is the width of the 95% confidence
interval?. Let z0.025 = 1.96.
Answer: We known σ = 1000, x̄ = 21000, n = 64, α = 0.05. The

with of the 95% confidence interval on mean is
σ 1000
2zα/2 √ = 2 ∗ z0.025 √ = 490
n 64

Known
Note
The length of a confidence interval is a measure of the precision of

estimation. From the preceeding discussion, we see that precision is
inversely related to the confidence level. It is desirable to obtain a
confidence interval that is short enough for decision-making purposes
and that also has adequate confidence. One way to achieve this is by
choosing the sample size n to be large enough to give a CI of specified
length or precision with prescribed confidence.

1.2 Choice of Sample Size
Sample Size for Specified Error on the Mean, Variance Known
If x̄ is used as an estimate of µ, we can be 100(1 − α)% confident that the

error | x̄ − µ | will not exceed a specified amount E when the sample size is
2
zα/2 σ
n=
E
√
2E = 2zα/2 σ/ n is the length of the resulting confidence interval.

Sample Size for Specified Error on the Mean,
Variance Known
Example 3
A nurse at a local hospital is interested in estimating the birth

weight of infants. How large a sample must she select if she desires
to be 98% confident that the true mean is within 4 ounces of the
sample mean? The standard deviation of the birth weights is known
to be 6 ounces. Let z0.01 = 2.33.

Sample Size for Specified Error on the Mean,
Variance Known
Example 3
A nurse at a local hospital is interested in estimating the birth

weight of infants. How large a sample must she select if she desires
to be 98% confident that the true mean is within 4 ounces of the
sample mean? The standard deviation of the birth weights is known
to be 6 ounces. Let z0.01 = 2.33.
Answer : We known σ = 6, E = 4, α = 0.02. Thus, the required

sample size:
2 2
z0.01 σ 2.33 × 6
n= = = 12.215 ≈ [12.215] = 13
E 4

1.3 One-Sided Confidence Bounds
One-Sided Confidence Bounds on the Mean, Variance Known
A 100(1 − α)% upper-confidence bound for µ is

√
µ ≤ u = x̄ + zα σ/ n
and a 100(1 − α)% lower-confidence bound for µ is

√
x̄ − zα σ/ n = l ≤ µ


√
µ ≤ u = x̄ + zα σ/ n

√
x̄ − zα σ/ n = l ≤ µ
Example 4
The diameter of holes for a cable harness is known to have a normal distri-
bution with σ = 0.01 inch. A random sample of size 15 yields an average
diameter of 1.5 inch. Find a 98% upper-confidence bound for the popula-
tion mean. Let z0.02 = 2.05.


√
µ ≤ u = x̄ + zα σ/ n

√
x̄ − zα σ/ n = l ≤ µ
Example 4
The diameter of holes for a cable harness is known to have a normal distri-
bution with σ = 0.01 inch. A random sample of size 15 yields an average
diameter of 1.5 inch. Find a 98% upper-confidence bound for the popula-
tion mean. Let z0.02 = 2.05.
Answer: We known σ = 0.01, x̄ = 1.5, α = 0.02, n = 15. Then a√98% upper-
confidence bound for µ is: µ ≤ x̄+zα σ/√n = 1.5+(2.05×0.01)/ 15 = 1.495
1.4 Large-Sample Confidence Interval for µ
Large-Sample Confidence Interval on the Mean
When n is large, the quantity

X̄ − µ
√
S/ n
has an approximate standard normal distribution. Consequently,

s s
x̄ − zα/2 √ ≤ µ ≤ x̄ + zα/2 √
n n
is a large sample confidence interval for µ, with confidence level of

approximately 100(1 − α)%.

Example 5
A civil engineer is analyzing the compressive strength of concrete. Com-

pressive strength is normally distributed. A random sample of 49 specimens
has a mean compressive strength of 3250(psi) and a variance 25(psi)2.
Construct approximately a 95% two-sided confidence interval on mean
compressive strength? Let z0.025 = 1.96

Example 5
A civil engineer is analyzing the compressive strength of concrete. Com-

pressive strength is normally distributed. A random sample of 49 specimens
has a mean compressive strength of 3250(psi) and a variance 25(psi)2.
Construct approximately a 95% two-sided confidence interval on mean
compressive strength? Let z0.025 = 1.96
Answer: We known n = 49, x̄ = 3250, s2 = 25, α = 0.05. Hence, a 95%
two-sided confidence interval on mean is
5 5
3250 − 1.96 √ ≤ µ ≤ 3250 + 1.96 √
49 49
3248.6 ≤ µ ≤ 3251.4

Question 1

Question 2

Question 3

Question 4

Question 5

Question 6

Question 7

Question 8

Question 9

Question 10

Question 11

Question 12

Question 13

Question 14

Question 15

Question 16

Question 17

Question 18

Question 19

Question 20

Question 21

Question 22

Question 23

Question 24

Question 25

Question 26

Content

Variance Known
Variance Unknown
Proportion

2. Confidence Interval on the Mean of a Normal
Distribution, Variance Unknown
When the population standard deviation σ is not know:

- n is reasonably large (n ≥ 30), we can even handle the case of unknown
variance for the large-sample-size situation in Section 1.4.


- However, when the sample is small:
We need to replace the population standard deviation σ by a sample
standard deviation s.


- However, when the sample is small:
We need to replace the population standard deviation σ by a sample
standard deviation s.
Use t-distribution instead of normal distribution.

What is t-distribution?
The t probability density function is

Γ[(k + 1)/2] 1
f (x) = √ . 2 , −∞ < x < ∞
πkΓ(k/2) [(x /k) + 1](k+1)/2
where, Γ(s) = R e−t ts−1 dt, s > 0, k is the number of degrees of freedom.
∞
The mean and variance of the t distribution are zero and k/(k − 2) (for
0
k > 2), respectively.

2.1 t Distribution
t Distribution
Let X1 , X2 , ..., Xn is a random sample from a normal distribution with

unknown mean µ and unknown variance σ2. The random variable
X̄ − µ
T = √
S/ n
has a t distribution with n − 1 degrees of freedom.

2.1 t Distribution
Let tα,k be the value of the random variable T with k degrees of freedom
above which we find an area (or probability) α. Then, t1−α,k = −tα,k .
Appendix Table V provides percentage points of the t distribution.
To illustrate the use of the table, note that the t-value with 10 degrees of
freedom having an area of 0.05 to the right is t0.05,10 = 1.812. That is,
P (T > t0.05,10 ) = P (T > 1.812) = 0.05
Therefore, t0.95,10 = −t0.05,10 = −1.812

2.2 t Confidence Interval on µ
X̄−µ
We know that the distribution of T = S/ √ is t with n − 1 degrees
n
of freedom. Letting tα/2,n−1 be the upper 100α/2 percentage point
of the t distribution with n − 1 degrees of freedom, we may write
P (−tα/2,n−1 < T < tα/2,n−1 ) = 1 − α
Hence,
S S
P (X̄ − tα/2,n−1 √ ≤ µ ≤ X̄ − tα/2,n−1 √ ) = 1 − α
n n

Confidence Interval on the Mean, Variance Unknown
If x̄ and s are the mean and standard deviation of a random sample from a
normal distribution with unknown variance σ2, a 100(1 − α)% confidence
interval on µ is given by
s s
x̄ − tα/2,n−1 √ ≤ µ ≤ x̄ + tα/2,n−1 √
n n
where tα/2,n−1 is the upper 100α/2 percentage point of the t distribution

with n − 1 degrees of freedom.


s
µ ≤ x̄ + tα,n−1 √
n

s
x̄ − tα,n−1 √ ≤ µ
n

Example 6
The brightness of a television picture tube can be evaluated by measuring

the amount of current required to achieve a particular brightness level. A
sample of 12 tubes results in x̄ = 316.8 and s = 15.2. Find (in microamps)
a 99% confidence interval on mean current required. Let t0.005,11 = 3.106

Example 6
The brightness of a television picture tube can be evaluated by measuring

the amount of current required to achieve a particular brightness level. A
sample of 12 tubes results in x̄ = 316.8 and s = 15.2. Find (in microamps)
a 99% confidence interval on mean current required. Let t0.005,11 = 3.106
Answer: We known n = 12, x̄ = 316.8, s = 15.2, α = 0.01 (variance un-
known model). A 99% confidence interval on mean is
15.2 15.2
316.8 − 3.106 √ ≤ µ ≤ 316.8 + 3.106 √
12 12
303.171 ≤ µ ≤ 330.429

Question 1

Question 2

Question 3

Question 4

Question 5

Question 6

Question 7

Question 8

Question 9

Question 10

Content

Variance Known
Variance Unknown
Proportion

3. Large-Sample Confidence Interval for a
Population Proportion
A random sample of size n has been taken from a large (possibly infinite)
population and that X(≤ n) observations in this sample belong to a class
of interest. Then P̂ = Xn is a point estimator of the proportion of the
population p that belongs to this class. Note that n and p are the parameters
of a binomial distribution. When n large, we known that
X − np P̂ − p
Z=p =p
np(1 − p) p(1 − p)/n
is approximately standard normal.

To construct the confidence interval on p, note that
P (−zα/2 ≤ Z ≤ zα/2 ) ∼
=1−α

so
p(1 − p)/n) ∼
p p
P (P̂ − zα/2 p(1 − p)/n ≤ p ≤ P̂ + zα/2 =1−α
The quantity is called the standard

p
p(1 − p)/n error of the point
estimator P̂ .
Approximate Confidence Interval on a Binomial Proportion
If p̂ is the proportion of observations in a random sample of size n that be-

longs to a class of interest, an approximate 100(1 − α)% confidence interval
on the proportion p of the population that belongs to this class is
p p
p̂ − zα/2 p̂(1 − p̂)/n ≤ p ≤ p̂ + zα/2 p̂(1 − p̂)/n
where zα/2 is the upper α/2 percentage point of the standard normal dis-
tribution.
Since P̂ is the point estimator of p, we can define the error in estimating p

by P̂ as E =| p− P̂ |. Note thatpwe are approximately 100(1−α)% confident
that this error is less than zα/2 p(1 − p)/n. In situations where the sample
size can be selected, we may choose n to be 100(1 − α)% confident that the
error is less than some specified value E.
Sample Size for a Specified Error on a Binomial Proportion
If we set E and solve for n, the appropriate sample

p
= zα/2 p(1 − p)/n
size is: zα/2 2
n=( ) p(1 − p)
E
Else, since max p(1 − p) = 0.25, we are at least 100(1 − α)% confident:
zα/2 2
n=( ) × 0.25
E

Example 7
Of 1000 randomly selected cases of lung cancer, 750 resulted in death

within 10 years.
a) Calculate a 95% two-sided confidence interval on the death rate from
lung cancer.
b) What sample size is needed to be 95% confident that the error in
estimating the true value of p is less than 0.04? Let z0.025 = 1.96.

Example 7
Of 1000 randomly selected cases of lung cancer, 750 resulted in death

within 10 years.
a) Calculate a 95% two-sided confidence interval on the death rate from
lung cancer.
b) What sample size is needed to be 95% confident that the error in
estimating the true value of p is less than 0.04? Let z0.025 = 1.96.
Answer. We known n = 1000, p̂ = 750/1000 = 0.75.
a) α = 0.05. A 95% two-sided confidence interval on the death rate from
lung cancer is
p p
(0.75−1.96 0.75 ∗ 0.25/1000; 0.75+1.96 0.75 ∗ 0.25/1000) = (0.723, 0.777)

Example 7
Answer. We known n = 1000, p̂ = 750/1000 = 0.75.

b) E = 0.04, α = 0.05. The required sample size
1.96 2
n=( ) × 0.75 × 0.25 ≈ [450.19] = 451
0.04

Example 8
A local men's clothing store is being sold. The buyers are trying to estimate
the percentage of items that are outdated. They will randomly sample
among its 9800 items in order to determine the proportion of merchandise
that is outdated. The current owners have never determined their outdated
percentage and can not help the buyers. Approximately how large a sample
do the buyers need in order to insure that they are 96% confident that the
error is within 2%? Let z0.02 = 2.05.

Example 8
A local men's clothing store is being sold. The buyers are trying to estimate
the percentage of items that are outdated. They will randomly sample
among its 9800 items in order to determine the proportion of merchandise
that is outdated. The current owners have never determined their outdated
percentage and can not help the buyers. Approximately how large a sample
do the buyers need in order to insure that they are 96% confident that the
error is within 2%? Let z0.02 = 2.05.
Answer: E = 0.02, α = 0.04. The required sample size is
2.05 2
n=( ) × 0.25 ≈ [2626.56] = 2627
0.02

Approximate OneSided Confidence Bounds on a Binomial
Proportion
The approximate 100(1 − α)% lower and upper confidence bounds are
p̂ − zα p̂(1 − p̂)/n ≤ p and p ≤ p̂ + zα p̂(1 − p̂)/n
p p

Question 1

Question 2

Question 3

Question 4

Question 5

Question 6

Question 7

Question 8

Question 9

Question 10

Question 11

Question 12

Question 13

Question 14

Question 15

Question 16

Question 17

Question 18

Question 19

Question 20

Question 21

Question 22

Chapter 7. Statistical Intervals For A Single Sample

Uploaded by

Copyright:

Available Formats

Chapter 7. Statistical Intervals For A Single Sample

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 7. Statistical Intervals For A Single Sample

Uploaded by

Copyright:

Available Formats

Chapter 7: Statistical Intervals for a Single

Course Name: PROBABILITY & STATISTICS

1 / 91 Chapter 7: Statistical Intervals for a Single Sample

1 Confidence Interval on the Mean of a Normal Distribution,

2 / 91 Chapter 7: Statistical Intervals for a Single Sample

In the previous chapter we illustrated how a parameter can be estimated

3 / 91 Chapter 7: Statistical Intervals for a Single Sample

1 Confidence Interval on the Mean of a Normal Distribution,

4 / 91 Chapter 7: Statistical Intervals for a Single Sample

Its Basic Properties

is a standard normal distribution.

A confidence interval (CI) estimate for µ is an interval of the form l ≤

Its Basic Properties

The endpoints or bounds l and u are called lower- and

6 / 91 Chapter 7: Statistical Intervals for a Single Sample

Its Basic Properties

Confidence Interval on the Mean, Variance Known

If x̄ is the sample mean of a random sample of size n from a normal

7 / 91 Chapter 7: Statistical Intervals for a Single Sample

The situation is illustrated in following figure, which shows several

8 / 91 Chapter 7: Statistical Intervals for a Single Sample

In a sample of 36 randomly selected women, it was found that their

9 / 91 Chapter 7: Statistical Intervals for a Single Sample

In a sample of 36 randomly selected women, it was found that their

Answer: We known σ = 2.5, x̄ = 65.3, n = 36, α = 0.1. Therefore,

(65.3 − 0.683, 65.3 + 0.683) = (64.617, 65.983)

9 / 91 Chapter 7: Statistical Intervals for a Single Sample

An economist is interested in studying the incomes of consumers

10 / 91 Chapter 7: Statistical Intervals for a Single Sample

An economist is interested in studying the incomes of consumers

Answer: We known σ = 1000, x̄ = 21000, n = 64, α = 0.05. The

10 / 91 Chapter 7: Statistical Intervals for a Single Sample

The length of a confidence interval is a measure of the precision of

11 / 91 Chapter 7: Statistical Intervals for a Single Sample

Sample Size for Specified Error on the Mean, Variance Known

If x̄ is used as an estimate of µ, we can be 100(1 − α)% confident that the

12 / 91 Chapter 7: Statistical Intervals for a Single Sample

A nurse at a local hospital is interested in estimating the birth

13 / 91 Chapter 7: Statistical Intervals for a Single Sample

A nurse at a local hospital is interested in estimating the birth

Answer : We known σ = 6, E = 4, α = 0.02. Thus, the required

13 / 91 Chapter 7: Statistical Intervals for a Single Sample

One-Sided Confidence Bounds on the Mean, Variance Known

A 100(1 − α)% upper-confidence bound for µ is

and a 100(1 − α)% lower-confidence bound for µ is

14 / 91 Chapter 7: Statistical Intervals for a Single Sample

One-Sided Confidence Bounds on the Mean, Variance Known

A 100(1 − α)% upper-confidence bound for µ is

and a 100(1 − α)% lower-confidence bound for µ is

14 / 91 Chapter 7: Statistical Intervals for a Single Sample

One-Sided Confidence Bounds on the Mean, Variance Known

A 100(1 − α)% upper-confidence bound for µ is

and a 100(1 − α)% lower-confidence bound for µ is

Large-Sample Confidence Interval on the Mean

When n is large, the quantity

has an approximate standard normal distribution. Consequently,

is a large sample confidence interval for µ, with confidence level of

15 / 91 Chapter 7: Statistical Intervals for a Single Sample