Applied Statistics and

Probability for Engineers

Sixth Edition
Douglas C. Montgomery George C. Runger
Covered Topics:
8-1 Confidence Interval on the Mean of a Normal distribution
Variance Known
8-1.1 Development of the Confidence Interval & Its Properties
8-1.2 Choice of Sample Size
8-1.3 One-Sided Confidence Bounds
8-1.4 Large-Sample Confidence Interval for μ

8-2 Confidence Interval on the Mean of a Normal distribution

Variance Unknown
8-2.1 t- Distribution
8-2.2 Confidence Interval on μ

8-3 Confidence Interval on the Variance & σ of a Normal

distribution
Learning Objectives for Chapter 8
After careful study of this chapter, you should be able to
do the following:
1. Construct confidence intervals on the mean of a normal distribution,
using normal distribution or t-distribution method.
2. Construct confidence intervals on variance and standard deviation
of normal distribution.
3. Construct confidence intervals on a population proportion.
4. Constructing an approximate confidence interval on a parameter.
 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.

 Information about the precision of estimation is conveyed by the

length of the interval

A short interval implies precise estimation

4
Estimation Process

Random Sample I am 95%

confident that
μ is between
Population 40 & 60
Mean
(mean, μ, is X = 50
unknown)

Sample

5
8-1 Confidence Interval on the Mean of a
Normal Distribution, Variance Known
8-1.1 Development of the Confidence Interval
and its Basic Properties

6
8-1.1 Confidence Interval and its Properties
 A two sided confidence interval estimate for µ is an interval of the form
𝒍𝒍 ≤ 𝝁𝝁 ≤ 𝒖𝒖,
where the end-points 𝒍𝒍 and 𝒖𝒖 are computed from the sample data

 Because different samples will produce different values of 𝒍𝒍 and 𝒖𝒖, these
values are random variables.

 Suppose that we can determine values of L and U such that the

following probability statement is true;

𝑃𝑃(𝐿𝐿 ≤ 𝜇𝜇 ≤ 𝑈𝑈) = 1 − α
• Where 0 ≤ α ≤ 1.
• There is a probability of 1- α of
selecting a sample for which the /2 1-α /2
confidence interval (CI) will
contain the true value of µ.
L µ U 7
8-1.1 Confidence Interval and its Properties (cont.)
 The endpoints or bounds L and U are called lower- and upper-
confidence limits, respectively.
 (𝟏𝟏 − 𝜶𝜶): is called the confidence coefficient
 Since Z follows a standard normal distribution, we can write:

Simplify:

8
Confidence Interval and its Properties

� is the sample mean of a random sample of size n from a normal

 If X
population with known variance σ2, a 100(1 − α)% CI on µ is given
by

x − zα / 2 σ / n ≤ µ ≤ x + zα / 2 σ / n

Where:
 zα/2 is the upper 100α/2 percentage point of the standard
normal distribution
 zα/2 is the normal distribution value for a probability of α/2 in each
tail

9
Common Levels of Confidence

Commonly used confidence levels are 90%, 95%, and 99%

Confidence
Confidence
Coefficient, Zα/2 value
Level
1-α
80% .80 1.28
90% .90 1.645
95% .95 1.96
98% .98 2.33
99% .99 2.58
99.8% .998 3.08
99.9% .999 3.27

10
8-1 Confidence Interval on the Mean of a
Normal Distribution, Variance Known
Definition

11
Example 8-1
Ten measurement of impact energy (J) on specimens of A238 steel cut at
60°C are as follows: 64.1, 64.7, 64.5., 64.6, 64.3, 64.5, 64.6, 64.8, 64.2,
and 64.3. Assume that impact energy is normally distributed with σ=1.
Find a 95% CI for µ, the mean impact energy !?

Solution:
 The required quantities are:
zα/2 = z0.025=1.96, n=10, σ=1 and mean 𝑥𝑥̅ = 64.46.
 The resulting 95% CI is found from equation 8.7 as follows:
σ σ
x − zα/2≤ µ ≤ x + zα / 2
n n
1 1
64.46 − 1.96 ≤ µ ≤ 64.46 + 1.96
10 10
63.84 ≤ µ ≤ 65.08

Interpretation: Based on the sample data, a range of highly plausible

values for mean impact energy for A238 steel at 60°C is
63.84J ≤ µ ≤ 65.08J
Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 12
Example 8-1 (cont.)
Interpreting a Confidence Interval
• The confidence interval is a random interval

Repeated construction of a confidence interval for µ.

Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 13

8-1 Confidence Interval on the Mean of a
Normal Distribution, Variance Known
 If an infinite number of random samples are collected and a
100(1-α)% confidence interval for µ is computed from each
sample:
 Thus, 100(1-α) % of these intervals will contain the true
value of µ.

 In practice we develop one confidence interval so we can define

the CI as following:
 The observed interval [𝐿𝐿, 𝑈𝑈] brackets the true value of µ,
with confidence 100(1-α).

Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 14

8-1 Confidence Interval on the Mean of a
Normal Distribution, Variance Known
Confidence Level and Precision of Error
The length of a confidence interval is a measure of the
precision of estimation.

Error in estimating µ with 𝑥𝑥̅

The larger the sample size (n) the smaller the error E
The largest value of E is 𝑧𝑧𝛼𝛼⁄2 𝜎𝜎/ 𝑛𝑛
Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 15
8-1 Confidence Interval on the Mean of a
Normal Distribution, Variance Known

8-1.2 Choice of Sample Size

Definition

E represents the length of the interval.

16
EXAMPLE 8-2: Metallic Material Transition
Consider the test described in Example 8-1.
Determine how many specimens must be tested to ensure that the
95% CI on 𝜇𝜇 for A238 steel cut at 60°C has a length of at most 1.0J.
Solution:
 The bound on error in estimation E is one-half of the length
of the CI.
 Use Eq. 8.8, to find 𝒏𝒏 that the 95% CI on µ has a length of
1.0 J With error E=0.5, σ = 1, and zα/2 = 1.96.

( )
2
 zα / 2 σ  2
 =  
1. 96 1
n =   = 15.37
 E   0.5 

 Since, n must be an integer, required sample size is n = 16.

Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 17
8-1.3 One-Sided Confidence Bounds
Definition

Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known

18
 Example 8-3 One-Sided Confidence Bound
The same data for impact testing from Example 8-1 are
used to construct a lower, one-sided 95% confidence
interval for the mean impact energy.
Solution:
Recall that zα = 1.64, n = 10, σ = l, and .

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

σ
x − zα ≤µ
n
1
64.46 −1.64 ≤µ
10
63.94 ≤ µ
Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 19
8-1.5 A Large-Sample Confidence Interval for µ
Definition
When µ and σ are unknown and the sample size (n) is
large we can replace σ with S in the CI equation with little
effect on the distribution of Z.

Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 20

Example 8-5 Mercury Contamination
A sample of fish was selected from 53 Arab Gulf locations
and mercury concentration in the muscle tissue was
measured (ppm). The mercury concentration values are:

1.230 1.330 0.040 0.044 1.200 0.270

0.490 0.190 0.830 0.810 0.710 0.500
0.490 1.160 0.050 0.150 0.190 0.770
1.080 0.980 0.630 0.560 0.410 0.730
0.590 0.340 0.340 0.840 0.500 0.340
0.280 0.340 0.750 0.870 0.560 0.170
0.180 0.190 0.040 0.490 1.100 0.160
0.100 0.210 0.860 0.520 0.650 0.270
0.940 0.400 0.430 0.250 0.270

Find an approximate 95% CI on µ.

Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 21

Example 8-5 Mercury Contamination (cont.)
The summary statistics for the data are as follows:

Because n > 40, the assumption of normality is not necessary to use in Equation
8-13. The required values are 𝑛𝑛 = 53, �𝑥𝑥 = 0.525, 𝑠𝑠 = 0.3486, and z0.025 = 1.96.
The approximate 95% CI on µ is
s s
x − z0.025 ≤ μ ≤ x + z0.025
n n
0.3486 0.3486
0.5250 − 1.96 ≤ μ ≤ 0.5250 + 1.96
53 53
0.4311 ≤ μ ≤ 0.6189
Interpretation: This interval is fairly wide because there is variability in the
mercury concentration measurements. A larger sample size would have
produced a shorter interval.
Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 22
Example 8-5 Mercury Contamination (cont.)

Figure 8-3 Mercury concentration in largemouth bass

(a) Histogram. (b) Normal probability plot
Sec 8-1 Confidence Interval on the Mean of a Normal, σ2 Known 23
8-2 Confidence Interval on the Mean of a
Normal Distribution, Variance Unknown
8-2.1 The t distribution

24
8-3 Confidence Interval on the Mean of a
Normal Distribution, Variance Unknown
8-3.1 The t distribution

n = 10

𝑛𝑛 = ∞ [𝑁𝑁 0,1 ]

n=1

Figure 8-4 Probability density functions of several t distributions.

25
8-3 Confidence Interval on the Mean of a
Normal Distribution, Variance Unknown
8-3.1 The t distribution

𝑡𝑡1−𝛼𝛼,𝑛𝑛 = −𝑡𝑡𝛼𝛼,𝑛𝑛 𝑡𝑡𝛼𝛼,𝑛𝑛

Figure 8-5 Percentage points of the t distribution.

26
8-3 Confidence Interval on the Mean of a
Normal Distribution, Variance Unknown
8-3.2 The t Confidence Interval on µ

One-sided confidence bounds on the mean are found by replacing

tα/2,n-1 in Equation 8-18 with t α,n-1.

27
 Example 8-6: Alloy Adhesion
Results of tensile adhesion test on 22 U-700 alloy specimens. The load at
specimen failure in mega pascals (Mpa):
19.8 10.1 14.9 7.5 15.4 15.4
15.4 18.5 7.9 12.7 11.9 11.4
11.4 14.1 17.6 16.7 15.8
19.5 8.8 13.6 11.9 11.4

 We want to find a 95% CI on µ.

 The sample mean is 𝑥𝑥̅ = 13.71 and sample standard deviation is s = 3.55.
 Since n = 22, we have n − 1 = 21 degrees of freedom for t, so t0.025,21 =
2.080.
 The resulting CI is x − t α/2, n −1s / n ≤ µ ≤ x + t α/2, n −1s/ n
13.71 − 2.080(3.55) / 22 ≤ µ ≤ 13.71 + 2.080(3.55) / 22
13.71 − 1.57 ≤ µ ≤ 13.71 + 1.57
12.14 ≤ µ ≤ 15.28

Interpretation: The CI is fairly wide because there is a lot of variability in the

tensile test measurements. A larger sample size would have led to a shorter
interval.
Sec 8-2 Confidence Interval on the Mean of a Normal, σ2 Unknown 28
Example 8-6 Alloy Adhesion

Figure 8-6. Box and Whisker plot Figure 8-7 Normal probability plot
for the load at failure data in of the load at failure data in
Example 8-6.
Example 8-6.

29
EXERCICE:

1) For a normal population with known 𝑉𝑉𝑉𝑉𝑉𝑉 𝑥𝑥 = 𝜎𝜎 2 . What is the

2.49𝜎𝜎
confidence level for the interval 𝑥𝑥̅ − ≤ 𝜇𝜇 ≤ 𝑥𝑥̅ + 2.49𝜎𝜎/ 𝑛𝑛
𝑛𝑛

2) For a normal population with known variance σ2 . What value of zα/2 in

the following equation
zα σ z𝛼𝛼 σ
x� − 2 ≤ μ ≤ x� + 2
n n
gives 98% confidence?

30