Escolar Documentos
Profissional Documentos
Cultura Documentos
CHAPTER 4 : ESTIMATION
Sub-Topic
Introduction.
Point estimation
Interval estimation.
Confidence interval for population mean.
Confidence interval for a difference between two means.
Confidence interval for a population variance.
Confidence interval for ratio of two variances.
Learning Objective
By the end of this chapter, students should be able to
Know how to construct confidence interval for population means and
variances.
Able to choose which distribution should be used in order to construct
confidence interval.
Able to choose appropriate sample size.
Bahasa Melayu
1.
Point estimation
Penganggar titik
2.
Interval estimation
Penganggar selang
3.
Aras keyakinan
4.
Confidence coefficient
Pekali keyakinan
153
Chapter 4: Estimation
4.1 Introduction
Definition 1
A point estimate is a single numerical value that used to estimates an unknown
population parameter.
Example 1
For example, a poll may seek to estimate the proportion of adult residents of a city
that support a proposition to build a new sports stadium. Out of a random sample of
200 people, 106 say they support the proposition. Thus in the sample, 0.53 of the
people supported the proposition. This value of 0.53 is called a point estimate of the
population proportion. It is called a point estimate because the estimate consists of a
single value or point.
154
Chapter 4: Estimation
Example 2
Suppose a random variable X is normally distributed with unknown population mean,
. After the sample has been selected, the numerical value of x is the point estimate
of . Thus, if the data are x1 25, x2 30, x3 29, x4 33, the point estimate of is
x
25 30 29 33
29.25
4
Definition 2
A confidence interval is a set of (real) numbers between two values that likely to
contain the parameter being estimated.
Definition 3
The confidence level of an interval estimate of a parameter is the probability that the
interval estimates will contain the estimated parameter. Point estimates are usually
supplemented by confidence intervals.
Theory 1
A confidence interval estimate for is an interval of the form l u . There is a
probability of (1 ) of constructing interval that will contain the true value of .
where
P{ l u }= 1
The relationship between and the confidence level is that the stated confidence
level is percentage equivalent to the decimal value of (1 ) .
Definition 4
The end-points or bounds l and u are called the lower-confidence limit and upperconfidence limits respectively.
155
Chapter 4: Estimation
Definition 5
(1 ) is called confidence coefficient.
Example 3
With 95% confidence interval, then 0.05 , since 1 0.05 0.95 .
When 0.01 , then 1 0.01 0.99 , and the 99% confidence interval is being
calculated.
Example 4
If the pollster used a method that contains the parameter 95% of the time it is used, he
or she would arrive at the following 95% confidence interval: 0.46 < < 0.60. The
pollster would then conclude that somewhere between 0.46 and 0.60 of the population
supports the proposal. The media usually reports this type of result by saying that
53% favor the proposition with a margin of error of 7%.
Theory 2
If the random variable X has a normal distribution with mean and variance 2 ,
then the sample mean X also has a normal distribution with mean , but with
variance 2 / n (refer previous chapter). In other words, X ~ N (
2
n
) . In fact, if we
have a random variable X which has any distribution (not necessarily normal), by the
central limit theorem the distribution of X will be approximately normal with mean
156
Chapter 4: Estimation
2 /n
Hence, P 1.96
1.96 0.95
/ n
x 1.96 / n x 1.96 / n
In general form given as :
x z / n x z / n
/2
(4.1)
/2
x z / 2 / n
or
(4.2)
If the value of is unknown or not given, so the above formula is used by substitute
x z / 2 s / n
(4.3)
The term z / 2 / n is called the maximum error of estimate, E. For specific value,
if 0.05 , 95% of the sample means will fall within this error value on either side
of the population mean. Refer Figure 1.
95%
0.025
z / 2 / n
Figure 1
157
z / 2 / n
0.025
Chapter 4: Estimation
Definition 6
The maximum error of estimate is the maximum likely difference between the point
estimate of a parameter and the actual value of the parameter.
Example 5
Suppose that 100 samples of water from a fresh water lake are taken and the calcium
concentration (milligrams per liter) is measured. The average is 0.66 mg/l and the
standard deviation is 0.049 mg/l. Construct 95% confidence interval for the
population mean.
Answer Example 5
n 100,
x 0.66, s 0.049
1 0.95 0.05
z / 2 z 0.05 / 2 z 0.025 1.96
x z / 2 s / n
0.66 0.0096
0.6504 0.6696
That is, based on the sample data, a range of the population mean of calcium
concentration from fresh water lake is between 0.6504 and 0.6696.
Theory 3
Sample size determination is closely related to statistical estimation. Quite often, one
asks, how large a sample is necessary to make an accurate estimate ? To determine
the minimum sample size for finding a confidence interval for the mean, the formula
for sample size is derived from the maximum error of estimate formula.
E Z 2
(4.4)
158
Chapter 4: Estimation
E n Z 2
Hence,
Z 2
E
Z 2
n
E
(4.5)
Example 6
A manufacturer is interested in the output voltage of a power supply used in a PC.
Output voltage is assumed to be normally distributed, with standard deviation 1.25V.
How large a sample must be selected if he wants to be 99% confident of finding
whether the true mean differs from the sample mean by 0.3V ?
Answer Example 6
z
n /2
2.5758(1.25)
0.3
115.1866
116
159
Chapter 4: Estimation
Theory 4
When 2 is unknown, a logical procedure is replace 2 with the sample variance s.
The random variable Z now becomes T
x
s2 / n
t distribution with n 1 degree of freedom. Hence the formula for mean population is
given by:
s
x t 2 , v
__
(4.6)
Example 7
The comprehensive strength of concrete is being tested by civil engineer. He tests 10
specimens and the obtained data presented in Table 1.
Table 1
2590
2530
2510
2566
2541
2557
2582
2550
2583
2599
Answer Example 7
x 2560 .8, s 28.5455, n 10
1 0.98 0.02
t / 2,v t 0.02 / 2,101 t 0.01,9 2.821
2560 .8 2.82128.5455 /
x t / 2,v s / n
10
2560 .8 25.4648
2535 .3352 2586 .2648
160
Chapter 4: Estimation
That is, based on the sample data, a range of the population mean strength of the
concrete is between 2535 .3352 and 2586 .2648 .
Example 8
A civil engineer is analyzing the compressive strength of concrete. A random sample
of 12 specimens has a mean compressive strength of 3201.33 psi and a standard
deviation of 900 psi. Construct a 95% confidence interval for the mean compressive
strength.
Answer Example 8
x 3201 .33, s 900, n 12
1 0.95 0.05
t / 2,v t0.05 / 2,121 t0.025,11 2.201
x t / 2,v s / n
3201.33 2.201 90 / 12
3201.33 57.1837
If all possible samples are taken and confidence interval estimates are
developed, 95% of them would include the true population mean
somewhere within their interval.
(b)
You have 95% confidence that you have selected a sample whose
interval does include the population mean.
(c)
b is true
(d)
161
Chapter 4: Estimation
2.
(b)
(c)
(d)
3.
4.
(a)
a narrower.
(b)
a wider.
(c)
the same.
(d)
a less significant.
Other things being equal, as the confidence level for a confidence interval
increase, the width of the interval increases.
5.
(a)
True.
(b)
False.
(b)
(c)
(d)
162
Chapter 4: Estimation
Subjective Questions
6.
(b)
(c)
If it was found that the sample mean is 10, find a 99% confidence
interval for the mean yield.
7.
8.
Suppose that in question 7 exercise 4.4 the sample size has been increased
from 20 to 50 inflators. Given that other measurements are still the same. Is
there any change for the confidence interval of the true mean ? If the answer is
yes, find the new confidence interval of the true mean.
9.
10.
A research engineer for a tire manufacturer is investigated tire life for new
rubber compound. Sample of 18 tires had been tested to end-of-life in a road
test and their average is 63,559 kilometers and standard deviation is 2,275
kilometers. Find a 99% confidence interval of the mean life tire.
163
Chapter 4: Estimation
11.
The college president asks the statistics teacher to estimate the average age of
the students at their college. How large a sample is necessary ? The statistics
teacher would like to be 99% confident that the estimate should be accurate
within one year. From a previous study, the standard deviation of the ages is
known to be 3 years.
12.
13.
A restaurant owner wishes to find the 99% confidence interval of the true
mean cost of a dry martini. How large should the sample be if she wishes to
be accurate within RM0.10 ? A previous study showed that the standard
deviation of the price was RM0.12.
14.
15.
Find the sample size needed to estimate the population mean to within one
fifth of a standard deviation with 99% confidence level.
2.
3.
4.
5.
164
Chapter 4: Estimation
Subjective Questions
6.
(a)
7.
(b)
60
(c)
(9,11)
(1.9916, 2.0484)
8.
(2.0036, 2.0364)
9.
(97.3747, 101.4253)
10.
(62,005.0268, 65,112.9732)
11.
60
12.
25
13.
10
14.
15.
167
Theory 5
In the previous chapter we already know that the mean sampling distribution of
12 2 2
.
X 1 X 2 is given by X 1 X 2 ~ N 1 2 ,
n1
n2
In the same way as confidence interval for population mean, we obtained the formula
for confidence interval for the difference between two means. Hence the confidence
interval in general form given by :
2
2
2
2
x1 x2 z / 2 1 2 1 2 x1 x2 z / 2 1 2
n1
n2
n1
n2
(4.7)
2 2
1
2
n1
n2
x1 x2 z / 2
or
(4.8)
We also decide whether to use the Z distribution or t distribution by using the same
principle.
If the sample size is large ( n 30 ) or known, we will use the Z
distribution.
If the sample size is small ( n 30 ) and unknown, we will use the t
distribution.
165
Chapter 4: Estimation
In other words, we only use the t distribution when the sample size is small ( n 30 )
and unknown, for other conditions we will use the Z distribution.
Z distribution case
When the value of is known, we will use (4.8) formula, but when is unknown
we simply substitute with the sample standard deviation, s. Thus from (4.8)
formula, we will obtained:
s2 s 2
1
2
n1 n2
x1 x2 z / 2
(4.9)
t distribution case
The general formula that will be used when we choose t distribution is given by :
s2 s 2
1
2
n1 n2
x1 x2 t / 2,v
(4.10)
Whereby, v n1 n2 2
We already know that the n1 30 , n2 30 and the value of 1 and 2 are unknown.
When we use t distribution, we need to take note on both sample size and the
information on the population standard deviations. The different information on these
will make the (4.10) formula varies. It can be summarize as follows:
Case (i) : n1 n2 , 21 2 2
Since n1 n2 , we can represent both sample size as n . So, n1 n2 n . From (4.10)
formula, we will obtained:
s2 s 2
1
2
n
n
x1 x2 t / 2,v
(4.11)
With v n1 n2 2 .
Or it can be simplify as :
1 2
2
s1 s 2
n
x1 x2 t / 2,v
166
(4.12)
Chapter 4: Estimation
With v 2n 2
Case (ii) : n1 n2 , 21 2 2
Since 21 2 2 , we can substitute both sample variances with pooled estimated
variance, S p2 . So, s 21 s 2 2 S p2 . From (4.10) formula, we will obtained :
S p2 S p2
x1 x2 t / 2,v
n
n
x1 x2 t / 2,v
Thus,
S p2
n
With v n n 2
Or it can be simplify as :
2
x1 x2 t / 2,v S p
(4.13)
With v 2n 2
S p is a pooled estimate of the variance, which given by :
S P2
Case (iii) : n1 n2 , 21 2 2
Since 21 2 2 , we can substitute both sample variances with pooled estimated
variance, S p2 . So, s 21 s 2 2 S p2 . From (4.9) formula, we will obtained :
S p2 S p2
x1 x2 t / 2,v
n1 n2
Or it can be simplify as :
167
(4.14)
Chapter 4: Estimation
1
1
n1 n2
x1 x2 t / 2,v S p
(4.15)
With v n1 n2 2
Case (iv) : n1 n2 , 21 2 2
To construct the confidence interval we still use the general formula in (4.10) :
s2 s 2
x1 x2 t / 2,v 1 2
n1 n2
s12 s22
n n2
v 12
2
s12
s22
n
1 n2
n1 1 n2 1
Example 9
A consumer organization collected data on two types of automobile batteries, A and
B. Both populations are normally distributed with standard deviations of 1.29 for
batteries A and 0.88 for batteries B. The summary statistics for 40 observations of
each type yielding average mean of 32.25 hours and 29.81 hours for batteries A and
batteries B respectively. Construct 90% confidence interval for difference between
means life hours for batteries A and batteries B.
Answer Example 9
A 1.29, B 0.88,
n A n B 40
x A 32.25,
x B 29.81
168
Chapter 4: Estimation
x A xB z / 2 A B
nA
nB
1.29 2 0.882
40
40
2.44 0.4061
2.0339 A B 2.8461
We are 90% confident that the difference between means life hours for batteries A
and batteries B is between 2.0339 and 2.8461 hours.
Example 10
The diameter of steel rods manufactured on two different extrusion machines is being
investigated. Two random samples of size n1 15 and n2 15 are selected, and the
sample means are 8.69cm and 8.51cm and sample variances are 0.30 and 0.44,
respectively from Machine 1 and Machine 2. Construct 95% confidence interval for
difference between means diameter of steel rods manufactured by Machine 1 and
Machine 2.
Answer Example 10
s 21 0.30, s 2 2 0.44,
n1 15, n2 15
x1 8.69,
x2 8.51
1 2
2
s1 s 2
n
x1 x2 t / 2,v
With v 2n 2 then, v 2(15) 2 28
169
Chapter 4: Estimation
1
0.30 0.44
15
0.18 0.4548
0.2748 1 2 0.6348
We are 90% confident that the difference between -0.2748 cm and 0.6348 cm.
Example 11
Reconsider the study on means diameter of steel rods in Example 10. If it is given
that the population variances are equal for both machines, construct the new
confidence interval for difference between means diameter of steel rods manufactured
by Machine 1 and Machine 2 at the same level of significant.
Answer Example 11
s 21 0.30, s 2 2 0.44,
n1 15, n2 15
x1 8.69,
x2 8.51
x1 x1 t / 2,v S p
With v 2n 2 then, v 2(15) 2 28
n1 1s12 n2 1s22
n1 n2 2
(15 1)0.30 (15 1)0.44
15 15 2
170
Chapter 4: Estimation
0.37
S p 0.6083
Substitute in (4.13) formula :
15
0.2749 1 2 0.6349
We are 90% confident that the difference between means diameter of steel rods
manufactured by Machine 1 and Machine 2 is between -0.2749cm and 0.6349cm.
Example 12
Reconsider the study on means diameter of steel rods in Example 10. If it is given
that the population variances are equal for both machines. Meanwhile the sample size
for Machine 1 has been changed from 15 to 10. Construct the new confidence interval
for difference between means diameter of steel rods manufactured by Machine 1 and
Machine 2 at the same level of significant.
Answer Example 12
s 21 0.30, s 2 2 0.44,
n1 10, n2 15
x1 8.69,
x2 8.51
x1 x 2 t / 2,v
1
1
S p
n
n
2
1
With v n1 n2 2 , then v 10 15 2
171
Chapter 4: Estimation
S P2
n1 1s12 n2 1s22
n1 n2 2
(10 1)0.30 (15 1)0.44
10 15 2
= 0.3852
S p 0.6206
Substitute in (4.15) formula :
15
0.2889 1 2 0.6489
We are 90% confident that the difference between means diameter of steel rods
manufactured by Machine 1 and Machine 2 is between -0.2889 cm and 0.6489 cm.
Example 13
Reconsider the study on means diameter of steel rods in Example 10. If it is given
that the population variances are not equal for both machines. Meanwhile the sample
size for Machine 1 has been changed from 15 to 10. Construct the new confidence
interval for difference between means diameter of steel rods manufactured by
Machine 1 and Machine 2 at the same level of significant.
Answer Example 13
s 21 0.30, s 2 2 0.44,
n1 10, n2 15
x1 8.69,
x2 8.51
172
Chapter 4: Estimation
s2 s 2
x1 x1 t / 2,v 1 2
n1 n2
Whereby,
2
1
n1 s 22 n 2
s 2 n
n1
2 2
n1 1
n2 1
2
1
15 1
21.7793
22
10
15
0.18 0.4986
0.3186 1 2 0.6786
We are 90% confident that the difference between means diameter of steel rods
manufactured by Machine 1 and Machine 2 is between -0.3186cm and 0.6786cm.
Exercise 4.5
1.
For the following problems, define whether the population standard deviations
are known or unknown. Then, choose the appropriate distribution in order to
construct the confidence interval for the difference in mean.
(a)
(b)
173
Chapter 4: Estimation
for each of two different levels of noise voltage, 100mV and 200mV.
For 100mV, the mean number of periods in a train was 7.9 with
variance of 4.7. Meanwhile, for 200mV, the mean number of periods
in a train was 6.5 with variance of 5.1
(c)
(d)
The overall distance traveled by a golf ball is tested by hitting the ball
with Iron Byron, a mechanical golfer with a swing that is said emulate
the legendry champion, Byron Nelson. 10 Randomly selected balls of
two different brands are tested and the overall distance measured. The
obtained data presented in Table 2.
Table 2
Brand A
275
286
287
260
277
273
269
282
281
275
Brand B
258
245
275
270
266
274
269
266
273
273
(e)
Mean
Standard deviation
Males
190
2.82
1.41
Females
220
2.42
1.30
174
Chapter 4: Estimation
2.
Refer to question 1(e) Exercise 5.5, given that the sample size for both male
and female have been reduced to 21. Construct a 90% confidence interval for
the difference between males and females in mean number of IT courses
planned to be taken in future.
3.
Before
60
71
63
59
75
70
68
After
57
69
56
51
72
66
62
Construct a 98% confidence interval for the difference weight before and after
the program.
4.
5.
175
Chapter 4: Estimation
Table 5
Sample Size
Mean
Variance
Type 1
13
173.5
3.7
Type 2
13
160.9
2.9
6.
7.
The deflection temperature under load for two different types of plastic pipe is
normally distributed with standard deviation of 16 C for Type A pipe and
21 C for Type B pipe. Two random samples of 13 pipe specimens from both
pipe types are tested and their means deflection temperatures are 125 C and
97 C respectively for Type A pipe and Type B pipe. Find a 99% confidence
interval for the difference in mean deflection temperatures between the two
types of pipe.
8.
A chemical engineer wish to know the difference mean yield between two
catalysts that have been used in a certain chemical process. Table 6 shows
data of yields from his experiment result.
Table 6
Catalyst X
80
88
76
83
85
85
81
79
75
Catalyst Y
77
85
81
76
81
74
79
80
83
176
Chapter 4: Estimation
Find a 99% confidence interval for the difference between the means yield
between two catalysts.
9.
10.
A study to see the difference between BMI of men and women was
conducted. A random sample of 11 men yielding average BMI of 28.9 and
standard deviation of 6.4. A random sample of 13 women yielding average
BMI of 26.1 and standard deviation of 4. Find a 99% confidence interval for
the difference between mean BMI for men and women.
(a)
known, Z distribution
(b)
unknown, t distribution
(c)
known, Z distribution
(d)
unknown, t distribution
(e)
unknown, Z distribution
2.
(-0.3048, 1.1048)
3.
(-5.1154, 14.544)
4.
(-73.627, -54.373)
5.
(10.6071, 14.5929)
6.
(-28.8678, -3.1323)
7.
(46.8607, 9.1393)
8.
(-3.6243, 7.1791)
9.
(-0.1135, 5.914)
10.
(-3.6505, 9.2505)
177
Chapter 4: Estimation
Theory 6
If s 2 is the sample variance from a random sample of n observations from a normal
distribution with unknown population variance 2 , a confidence interval on 2 is
given as :
(n 1) s 2
2 / 2,v
(n 1) s 2
(4.16)
12 / 2,v
Whereby 2 / 2,v and 12 / 2,v are the upper and lower 100 / 2 percentage points of the
chi-square distribution with v n 1 degree of freedom, respectively.
5.6.2 Confidence Interval for Ratio of Two Population Variances
Theory 7
If s12 and s 22 are the sample variances from a random samples of n 1 and n 2
respectively, from two independent normal populations with unknown population
variances 12 and 22 , a confidence interval on the ratio
1 s12
s12
1
f / 2 ,v 2 ,v1
s22 f / 2,v1,v 2 22 s22
2
is given as :
22
1
1
f / 2 ,v1,v 2
(4.17)
178
Chapter 4: Estimation
Example 14
The life hours of a heating element used in a furnace is known to be approximately
normally distributed. A random sample of 11 heating elements Type A is selected and
found to have sample standard deviation of 9.7 hours. Meanwhile a random sample of
13 heating elements Type B is selected and found to have sample standard deviation
of 11.4 hours. Construct a 95% confidence interval for the ratio population variances
A2
.
B2
Answer Example 14
n A 11, s A2 9.7 2
v A n A 1 11 1 10
v B nB 1 13 1 12
0.05, / 2 0.025
f / 2,vB,vA f 0.025,12,10 3.62
1
f / 2,vA,vB
1
f 0.025,10,12
1
3.37
1 s A2
s A2
1
f / 2,vB ,vA
sB2 f / 2,vA,vB 22 sB2
2
12 9.7 2
1
3.62
11.4 2 3.37 22 11.4 2
9.7 2
2
0.2148 2 2.6209
2
1
179
A2
is between 0.2148 and
B2
Chapter 4: Estimation
Example 15
A polymer is manufactured in a batch chemical process. A random sample of 9
viscosity measurements yielding standard deviation of 19. Following the process
change, 8 batch viscosity measurements yielding standard deviation of 16. Construct
a 98% confidence interval for the ratio population variances
12
.
22
Answer Example 15
n1 9, s`21 19 2
n2 8, s`22 16 2
v1 n1 1 9 1 8
v2 n2 1 8 1 7
0.02, / 2 0.01
f / 2,v 2,v1 f 0.01,7,8 6.18
1
f / 2,v1,v 2
1
f 0.01,8,7
1
6.84
1 s12
s12
1
f / 2 , v 2 ,v 1
s22 f / 2,v1,v 2 22 s22
2
19 2 1 12 19 2
6.18
16 2 6.84 22 16 2
0.2062
2
8.7148
22
1
180
A2
is between 0.2062 and
B2
Chapter 4: Estimation
Exercise 4.6
1.
2.
3.
4.
5.
6.
181
Chapter 4: Estimation
7.
8.
2
Before
2
After
9.
A2
B2
A fuel-economy study was conducted for two local automobiles, X and Y. One
vehicle of each brand was selected, and the mileage performance was
observed for 9 tanks of fuel in each car. From the study it was found that X
has a variance of 0.41 liter and Y has a variance of 0.34 liter. Construct a 95%
confidence interval for the ratio population variances.
10.
Table 7 shows the data of waiting times (in minutes) of customers at ABC
Bank and XYZ Bank. Construct a 90% confidence interval for the ratio
population variances.
Table 7
ABC Bank
6.1
5.8
7.0
6.4
6.9
7.3
XYZ Bank
6.8
5.4
7.4
7.1
6.6
6.0
182
6.5
Chapter 4: Estimation
11.
12.
Consider the deflection temperature under load for two different types of
plastic pipe described in question 6 Exercise 4.5. Compute a 98% confidence
interval for the ratio population variances,
13.
2
TypeA
2
Type
B
14.
X2
Y2
15.
Before
After
Consider the study to see the difference between BMI of men and women in
question 10 Exercise 4.5. Compute a 90% confidence interval for the ratio
population variances,
2
Male
2
Female
(15.1850, 39.9307)
2.
(0.006343, 0.0181)
3.
(8.8724, 42.0673)
4.
(1.3320, 13.6919)
5.
(0.0395, 0.0950)
6.
(0.5158, 1.3693)
7.
(0.0741, 5.3164)
8.
(0.2193, 2.5063)
183
Chapter 4: Estimation
9.
(0.2722, 5.3421)
10.
(0.1037, 2.253)
11.
(0.6051, 4.3788)
12.
(0.1686, 4.9384)
13.
(0.3519, 5.5311)
14.
(0.1635, 1.935)
15.
(0.9309, 7.4496)
EXERCISE CHAPTER 4
1.
A major truck stop has kept extensive records on various transactions with its
customers. If a random sample of 18 of these records shows average sales of
63.84 gallons of diesel fuel with a standard deviation of 2.75 gallons,
construct a 99% confidence intervals for the mean of the population sampled.
2.
3.
5900
6090
6310
7160
8440
9930
A lathe is set to cut steel bars and considered to be in perfect adjustment if the
average length of the bars it cuts is 7 centimeters. A sample of 31 bars is
selected randomly, and the lengths are measured. It is determined that the
average length of the bars in the sample is 7.055 centimeters with a standard
deviation of 0.35 centimeters. Find a 97% confidence interval that the average
length of the bars is in perfect adjustment.
4.
184
Chapter 4: Estimation
5.
To find an interval estimate of the number of miles a certain brand tires will
last before tread depth falls below minimal safety threshold, a manufacturer
tests 50 tires under various operating conditions. The recorded data yield the
estimates x 32,460 miles and s 3106 . Find an 80% confidence interval
for the mean number of miles.
6.
8.30
8.27
8.22
8.29
8.39
8.21
8.38
8.35
8.37
7.
(b)
(c)
(d)
Is there any difference between the result in question (b) and (c)?
Why ?
8.
185
Chapter 4: Estimation
of bacteria and were then observed daily. If the population variances of both
group are not equal, find the 99% confidence interval for the difference
between the average days of 2 groups of mice that infected by the dose of
bacteria.
Table 10
Mouse
Group
9.
2, 2, 3, 3, 3, 3, 4, 4, 5, 6, 7, 7, 8, 9
1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 5
10.
A sample of 14 cans of Brand I diet soda gave the mean number of calories of
23 per can with a standard deviation of 3 calories. Another sample of 16 cans
of Brand II diet soda gave the mean number of calories of 25 per can with a
standard deviation of 4 calories. Assume that the calories per can of diet soda
are normally distributed for each of the two brands and that the standard
deviations for the two populations are equal. Find the 99% confidence interval
for 1 2 .
11.
Pre training : 99 57 62 69 74 77 59 92 70 85 84
186
Chapter 4: Estimation
Post training : 94 57 62 69 66 76 58 88 70 84 83
Construct a 95% confidence interval for the difference between the weights in
pre training and the weights in post training by assuming 12 22 .
12.
Twelve randomly selected mature Jati trees of one variety have a mean height
of 13.8 feet with a standard deviation of 1.2 feet, and 15 randomly selected
mature Jati trees of another variety have a mean height of 12.9 feet with a
standard deviation of 1.5 feet. Assuming that the random samples were
selected from normal population with equal variances, construct a 95%
confidence interval for the difference between the true average height of the
two kinds of Jati trees.
13.
14.
A medical researcher wants to determine whether male pulse rates vary more
or less than female pulse rates. The statistics that he found from his research
can be summarized as shows in Table 11:
Table 11
Male
Female
Number of samples : 7
Number of samples : 9
Mean : 69.4
Mean : 76.3
(a)
(b)
15.
187
Chapter 4: Estimation
customers may enter any one of three different lines that have formed at three
teller windows.
Table 12
Single-line
system
Multiple-line
system
(a)
6.5
6.6
6.7
6.8
7.1
7.3
7.4
7.7
7.7
7.7
4.2
5.4
5.8
6.2
6.7
7.7
7.7
8.5
9.3
10.0
(b)
(c)
16.
(b)
17.
(b)
188
Chapter 4: Estimation
18.
45
Table 13
10
18
16
29
18
20
39
38
33
(a)
(b)
19.
Two groups of students are given a problem-solving test, and the results are
shown in Table 14. Construct the 95% confidence interval for the ratio of the
standard deviation for the two groups,
1
.
2
Table 14
Group 1 (Finance Major)
Group 2 (Management Major)
Sample size
Variance
20.
13
15.9
11.4
A study was performed to determine whether men and women differ in their
repeatability in assembling components on printed circuit boards. Two
samples of 15 men and 17 women were selected, and each subject assembled
the units. The two sample standard deviations of assembly time were 1.21
minutes and 1.35 minutes.
(a)
(b)
189
Chapter 4: Estimation
m2 en
times for men and women, 2
.
wom en
ANSWER EXERCISE CHAPTER 4
1.
(61.9616, 65.7184)
2.
(5552.120, 8530.735)
3.
(6.9186, 7.1914)
4.
(36.5104, 39.4895)
5.
(31897, 33023)
6.
(8.240, 8.362)
7.
(a)
0.12544
(b)
(0.990992, 1.349008)
(c)
(1.060496, 1.279504)
(d)
Yes.
8.
(0.1428, 4.1948)
9.
(0.9812, 1.1588)
10.
(-5.6104, 1.6104)
11.
(-9.7404, 13.5586)
12.
(-0.198, 1.998)
13.
(a)
(b)
(44.3625, 16.9966)
14.
(0.1757, 4.5764)
15.
(a)
(0.329, 0.870)
(b)
(1.253, 3.326)
(c)
(7.2817, 24.8868)
(a)
(0.308, 1.139)
(b)
17.
(a)
0.12544
(b)
(0.99092, 1.349008)
(c)
(0.99092, 1.349008)
18.
(a)
139.4242
(b)
(77.950, 335.228)
19.
(0.576, 2.213)
20.
(a)
(b)
(0.2398, 2.4663)
(0.2242, 1.0289)
190
(-0.33, 1.130)
Chapter 4: Estimation
SUMMARY CHAPTER 4
Confidence Interval for Single Mean
Z 2
Maximum error : E Z 2
, Sample size : n
n
E
(a)
(b)
(i)
is known : x z / 2 / n x z / 2 / n
(ii)
is unknown : x z / 2 s / n x z / 2 s / n
x t 2,v s / n x t 2,v s / n
; v n 1
(b)
Z distribution case
(i)
2 2
is known : x1 x2 z / 2 1 2
n1
n2
(ii)
s2 s 2
is unknown : x1 x2 z / 2 1 2
n1
n2
t distribution case
1 2
2
s1 s 2
n
(i)
n1 n 2 , 12 22 :
x1 x2 t / 2,v
(ii)
n1 n 2 , 12 22 :
x1 x2 t / 2,v S p
S P2
(iii)
; v 2n 2
2
; v 2n 2
1
1
n1 n 2 , 12 22 : x1 x 2 t / 2,v S p
; v n1 n2 2
n1 n2
191
Chapter 4: Estimation
S P2
(iv)
s2 s 2
n1 n 2 , 12 22 : x1 x2 t / 2,v 1 2 ,
n1 n2
(n 1) s 2
2 / 2,v
(n 1) s 2
12 / 2,v
; v n 1
1 s12
s12
1
192
s12 s22
n n2
v 12
2
s12
s22
n1 n2
n1 1 n2 1
Chapter 4: Estimation
193
Chapter 4: Estimation
194