Example: An electrical firm manufactures light bulbs that have a length of life that is approx-

imately normally distributed with a standard deviation of 40 hours. If a sample of 30 bulbs has
an average life of 780 hours, find a 96 percent confidence interval for the population mean of all
bulbs produced by this firm.
Solution: Here n = 30, σ = 40 and x = 780, Now to find α for 96 percent of confidence interval,
we have

100(1 − α) = 96
100 − 100α = 96
100α = 100 − 96
4
α = = 0.04
100
α
= 0.02.
2
The z-value leaving an area of 0.02 to the right, and therefore an area of 0.98 to the left. So
z0.02 = 2.05 (Table A.3). So
σ 40
x − z α2 √ = 780 − (2.05) √ = 765
n 30
σ 40
x + z α2 √ = 780 + (2.05) √ = 795
n 30
Hence, the 96 percent confidence interval is

765 < µ < 795.

Theorem 0.1. If x is used as an estimate of µ, we can be 100(1 − α) percent confident that the
error will not exceed z α2 √σn .

Example: In Example 1, we are 95 percent confident that the sample mean x = 2.6 differs
from the true mean µ by an amount less than (1.96)(0.3)
√
36
= 0.1 and 99 percent confident that the
(2.575)(0.3)
difference is less than √
36
= 0.13.
Example: The heights of a random sample of 50 college students showed a mean of 174.5 centime-
ters and a standard deviation of 6.9 centimeters. What can we assert with 98 percent confidence
about the possible size of our error if we estimate the mean height of all college students to be
174.5 centimeters?

5
Solution: Here σ = 6.9 and n = 50. To find α, we have

100(1 − α) = 98
100 − 100α = 98
100α = 100 − 98
2
α = = 0.02
100
α
= 0.01.
2
The z-value leaving an area of 0.01 to the right, and therefore an area of 0.99 to the left. So
z0.01 = 2.33 (Table A.3). So
σ (2.33)(6.9)
e < z α2 √ = √ = 2.27
n 50

Theorem 0.2. If x is used as an estimate of µ, we can be 100(1 − α) percent confident that the
error will not exceed a specified amount e when the sample size is
 z α σ 2
2
n=
e
Example: How large a sample is required if we want to be 95 percent confident that our
estimate of µ in Example 1 is off by less than 0.05?.
Solution: The population standard deviation is σ = 0.3. Then, by Theorem 0.2,
 
(1.96)(0.3)
n= = 138.3
0.05

Example: A UCLA researcher claims that the life span of mice can be extended by as much
as 25 percent when the calories in their diet are reduced by approximately 40 percent from the
time they are weaned. The restricted diet is enriched to normal levels by vitamins and protein.
Assuming that it is known from previous studies that σ = 5.8 months, how many mice should be
included in our sample if we wish to be 99 percent confident that the mean life span of the sample
will be within 2 months of the population mean for all mice subjected to this reduced diet?
Solution: Here σ = 5.8 and e = 2. Now to find α, we have

100(1 − α) = 99
100 − 100α = 99
100α = 100 − 99
1
α = = 0.01
100
α
= 0.005
2

6
The z-value leaving an area of 0.005 to the right, and therefore an area of 0.995 to the left. So
z0.005 = 2.57 (Table A.3). So
 2
(2.57)(5.8)
n= = 56 when rounded up
2

One-Sided Confidence Bounds:

The confidence intervals and resulting confidence bounds discussed thus far are two-sided (i.e.,
both upper and lower bounds are given). However, there are many applications in which only
one bound is sought. For example, if the measurement of interest is tensile strength, the engineer
receives better information from a lower bound only. On the other hand, if the measurement
is something for which a relatively large value of µ is not profitable or desirable, then an upper
confidence bound is of interest. An example would be a case in which inferences need to be made
concerning the mean mercury composition in a river. An upper bound is very informative in this
case.

One sided Confidence bounds on known µ and σ 2

If X is the mean of a random sample of size n from a population with variance σ 2 , the one-sided
100(1 − α) percent confidence bounds for µ are given by
σ
U pper one − sided bound : x + zα √
n
σ
Lower one − sided bound : x − zα √
n

Example: In a psychological testing experiment, 25 subjects are selected randomly and their
reaction time, in seconds, to a particular stimulus is measured. Past experience suggests that the
variance in reaction times to these types of stimuli is 4 sec2 and that the distribution of reaction
times is approximately normal. The average time for the subjects is 6.2 seconds. Give an upper
95 percent bound for the mean reaction time.
√
Solution: The point estimate of µ is x = 6.2 and σ = 4 = 2 and n = 25. Now to find α for 95
of confidence interval, we have

100(1 − α) = 95
100 − 100α = 95
100α = 100 − 95
5
α = = 0.05.
100
The z-value leaving an area of 0.05 to the right, and therefore an area of 0.95 to the left. So
z0.05 = 1.64 (Table A.3). So 95 percent upper bound is given by
r
σ 4
x + zα √ = 6.2 + (1.64) = 6.858.
n 25

7
Hence, we are 95 percent confident that the mean reaction time is less than 6.858 seconds.

Exercise: 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, 9.29, 9.31.

Estimating the difference between two Means:

If we have two populations with means µ1 and µ2 and variances σ12 and σ22 , respectively, then we
can find a point estimator of the difference between µ1 and µ2 .
Confidence interval for µ1 − µ2 , σ12 and σ22 known:
If x1 and x2 are means of independent random samples of sizes n1 and n2 from populations with
known variances σ12 and σ22 , respectively, a 100(1 − α) percent confidence interval for µ1 − µ2 is
given by s s
σ12 σ22 σ12 σ22
(x1 − x2 ) − z α2 + < µ1 − µ2 < (x1 − x2 ) + z α2 + ,
n1 n2 n1 n2
α
where z α2 is the z-value leaving an area of 2
to the right.

Example: A study was conducted in which two types of engines, A and B, were compared. Gas
mileage, in miles per gallon, was measured. Fifty experiments were conducted using engine type
A and 75 experiments were done with engine type B. The gasoline used and other conditions were
held constant. The average gas mileage was 36 miles per gallon for engine A and 42 miles per gallon
for engine B. Find a 96 confidence interval on µB − µA , where µA and µB are population mean
gas mileages for engines A and B, respectively. Assume that the population standard deviations
are 6 and 8 for engines A and B, respectively.
Solution: Here n1 = 50, n2 = 75, σ1 = 6, σ2 = 8, x1 = 36 and x2 = 42. Now as calculated
previously α = 0.04 and so α2 = 0.02 and z0.02 = 2.05. So
s r
σ12 σ22 82 62
(x2 − x1 ) − z α2 + = (42 − 36) − (2.05) + = 3.43
n1 n2 75 50
s r
σ12 σ22 82 62
(x2 − x1 ) + z α2 + = (42 − 36) + (2.05) + = 8.57
n1 n2 75 50

Hence a 96 confidence interval on µB − µA is

3.43 < µB − µA < 8.57.

Exercise: 9.35, 9.36.

Single Sample: Estimating the Proportion:

A point estimator of the proportion p in a binomial experiment is given by the statistic P = Xn ,
where X represents the number of successes in n trials. Therefore, the sample proportion p = nx
will be used as the point estimate of the parameter p.
Confidence Intervals for p:

8
If p is the proportion of successes in a random sample of size n and q = 1 − p, an approximate
100(1 − α) percent confidence interval, for the binomial parameter p is given by
r r
pq pq
p − z α2 < p < p + z α2 ,
n n
α
where z α2 is the z-value leaving an area of 2
to the right.

Example: In a random sample of n = 500 families owning television sets in the city of Hamilton,
Canada, it is found that x = 340 subscribe to HBO. Find a 95 percent confidence interval for the
actual proportion of families with television sets in this city that subscribe to HBO.
Solution: The point estimate of p is p = nx = 340
500
= 0.68 and so q = 1 − 0.680.32. Here α = 0.05
and hence z0.025 = 1.96. Therefore 95 percent confidence interval for p is
r
pq p
p − z α2 = 0.68 − (1.96) (0.68)(0.32)500 = 0.6391
n
r
pq p
p + z α2 = 0.68 + (1.96) (0.68)(0.32)500 = 0.7209.
n
Hence
0.6391 < p < 0.7209.

Theorem 0.3. If p is used q as an estimate of p, we can be 100(1 − α) percent confident that the
error will not exceed z α2 pnq .

Theorem 0.4. If p is used as an estimate of p, we can be 100(1 − α) percent confident that the
error will be less than a specified amount e when the sample size is approximately

z 2α p q
2
n = .
e2
Note: Do Example 9.15 of book.

Estimating the difference between two Proportions:

Confidence Interval for p1 − p2 :
If p1 and p2 are the proportions of successes in random samples of sizes n1 and n2 , respectively,
q 1 = 1 − p1 , and q 2 = 1 − p2 , an approximate 100(1 − α) percent confidence interval for the
difference of two binomial parameters, p1 − p2 , is given by
r r
p1 q 1 p2 q 2 p1 q 1 p2 q 2
(p1 − p2 ) − z α2 + < p1 − p2 < (p1 − p2 ) + z α2 + ,
n1 n2 n1 n2
α
where z α2 is the z-value leaving an area of 2
to the right.

9
Example: A certain change in a process for manufacturing component parts is being considered.
Samples are taken under both the existing and the new process so as to determine if the new
process results in an improvement. If 75 of 1500 items from the existing process are found to be
defective and 80 of 2000 items from the new process are found to be defective, find a 90 percent
confidence interval for the true difference in the proportion of defectives between the existing and
the new process.
Solution: Let p1 and p2 be the true proportions of defectives for the existing and new processes,
respectively. So
p1 75
p1 = = = 0.05
n1 1500
p2 80
p2 = = = 0.04
n2 2000

q 1 = 1 − p1 = 1 − 0.05 = 0.95

q 2 = 1 − p2 = 1 − 0.04 = 0.96

p1 − p2 = 0.05 − 0.04 = 0.01.

Now to find α for 90 percent of confidence interval, we have
100(1 − α) = 90
100 − 100α = 90
100α = 100 − 90
10
α = = 0.1
100
α
= 0.05
2
The z-value leaving an area of 0.05 to the right, and therefore an area of 0.95 to the left. So
z0.05 = 1.64 (Table A.3). So
r r
p1 q 1 p2 q 2 (0.05)(0.95) (0.04)(0.96)
(p1 − p2 ) − z α2 + = 0.01 − 1.64 + = − 0.0017
n1 n2 1500 2000
r r
p1 q 1 p2 q 2 (0.05)(0.95) (0.04)(0.96)
(p1 − p2 ) + z α2 + = 0.01 + 1.64 + = 0.0217
n1 n2 1500 2000
Hence

−0.0017 < p1 − p2 < 0.0217.

Example: A new rocket-launching system is being considered for deployment of small, short-
range rockets. The existing system has p = 0.8 as the probability of a successful launch. A sample
of 40 experimental launches is made with the new system, and 34 are successful.

10
 (1). Construct a 95 confidence interval for p.

(2). Would you conclude that the new system is better?.

Solution: Here p = nx = 40 34
= 0.85 and n = 40. So q = 1 − p = 1 − 0.85 = 0.15. We have α = 0.05
and hence z α2 = z0.025 = 1.96. Now

(1). 90 percent confidence interval is

r r
pq (0.85)(0.15)
p − z α2 = 0.85 − 1.96 = 0.739
n 40
r r
pq (0.85)(0.15)
p + z α2 = 0.85 + 1.96 = 0.961
n 40
Hence
0.739 < p < 0.961.

(2). Since p = 0.8 falls in the confidence interval, we cannot conclude that the new system is
better.

Example: A geneticist is interested in the proportion of African males who have a certain minor
blood disorder. In a random sample of 100 African males, 24 are found to be afflicted.

(1). Compute a 99 percent confidence interval for the proportion of African males who have this
blood disorder.

(2). What can we assert with 99 percent confidence about the possible size of our error if we
estimate the proportion of African males with this blood disorder to be 0.24?

Solution: Here p = nx = 10024

= 0.24 and n = 100. So q = 1 − p = 1 − 0.24 = 0.76. We have
α = 0.01 and hence z 2 = z0.005 = 2.57. Now Now
α

(1). 99 percent confidence interval is

r r
pq (24)(0.76)
p − z α2 = 0.24 − 2.57 = 0.130
n 100
r r
pq (24)(0.76)
p + z α2 = 0.24 + 2.57 = 0.350
n 100
Hence
0.130 < p < 0.350.

(2). Possible size of error will be

r r
pq (24)(0.76)
e ≤ z α2 = 2.57 = 0.110
n 100

11
Example: A random sample of 200 voters in a town is selected, and 114 are found to support
an annexation suit. How large a sample is needed if we wish to be 96 percent confident that our
sample proportion will be within 0.02 of the true fraction of the voting population?
x 114
Solution: Here α = 0.04 and hence z α2 = z0.02 = 2.05. Also p = n
= 200
= 0.57 and hence
q = 1 − p = 0.43. Also e = 0.02. So

z 2α p q (2.05)2 (0.57) (0.43)

2
n = = = 2576.‘
e2 (0.02)2

Exercise: 9.51, 9.52, 9.53, 9.54, 9.55, 9.56, 9.57, 9.59, 9.60, 9.61.

12