Chap 006

Chapter 06 - Confidence Intervals
CHAPTER 6
CONFIDENCE INTERVALS
6.1.
A confidence interval is an interval of values believed to contain an unknown population

parameter. It has a given level of our belief, or confidence, that the interval contains the unknown
parameter.
6.2.
Once the sampling has taken place, X is realized and, though unknown, it is either in the given
interval, or not in the interval. Nothing is random at this point and hence we cannot talk about
probabilities.
6.3.
If the pre-sampling probability that the parameter will be captured in an interval is 1 , then (1
)100% of the intervals that would be constructed this way would contain the unknown
parameter. Hence the confidence level (1 )100% that we attach to a given, single interval we
obtain.
6.4.
An expensive way of solving the problem: buy more information. As n increases, the width of
interval decreases.
Or, a cheap way to solve the problem: lower your required confidence level from 95% to 90% or
lower. As the confidence level decreases, the interval width decreases with it.
6.5.
95% C.I.:
x 1.96( /
n ) = 89,673.12
1.96(5,500/
16 )
= 89,673.12 2,695 = [86,978.12, 92,368.12] dollars.

Confidence Interval for
Known
Population Normal?
Population Stdev.
5500
Sample Size
16
n
Sample Mean 89673.12 x-bar
(1)
99%
95%
6.6.
Confidence Interval
89673.12 3541.77
89673.12 2694.95
=
=
[86131.3,93214.9
[86978.2,92368.1
99% C.I.:
89,673.12 2.576(5,500/ 16 ) = [86,131.12, 93,215.12]
As expected, this interval is wider than the 95% C.I. based on the same results.
6-1
6-7.
x = 32
= 4.6 n = 100
95% C.I. for :
32
1.96(4.6/
100 ) = 32
.9016 = [31.098, 32.902] m.p.g.
6-8.
We need not assume a normal population. n = 100 > 30, so the Central Limit Theorem applies.
6-9.
= 1.2
90% C.I.:
x = 9.3
n = 60
9.3 (1.645)
1.2
60
= [9.045, 9.555] percent alcohol.
(template : Estimating Mean.xls, sheet : sample stats)

Known
Population Normal?
Population Stdev.
Sample Size
Sample Mean
(1)
99%
95%
90%
1.2
60
9.3
n
x-bar
Confidence Interval
9.3 0.39905 = [ 8.90095,9.69905
9.3 0.30364 = [ 8.99636,9.60364
9.3 0.25482 = [ 9.04518,9.55482
]
]
]
6-10. = 430 n = 30 x = 1482

Known
Population Normal?
Population Stdev.
Sample Size
Sample Mean
(1)
99%
95%
430
30
1482
n
x-bar
Confidence Interval
1482 202.221
1482 153.87
6-2
=
=
[1279.78,1684.22]
[1328.13,1635.87]
x = 1700
= 800 n = 50
6.11.
Known
Population Normal?
Population Stdev.
Sample Size
Sample Mean
(1)
99%
95%
90%
n
x-bar
800
50
1700
Confidence Interval
1700 291.422
1700 221.744
1700 186.094
=
=
=
[1408.58,1991.42
[1478.26,1921.74
[1513.91,1886.09
The 90% C.I.: 1700 186.094 [1513.91, 1886.09]

b. The 90% C.I. contains the value of 1800. The airline should fly the route.
6-12.
= 55,000
n = 80
x = 165,000
80% C.I. for : 165000 7880.5, [157120, 172880]

Known
Population Normal?
Population Stdev.
Sample Size
Sample Mean
(1)
99%
95%
90%
80%
55000
80
165000
n
x-bar
Confidence Interval
165000 15839.3
165000 12052.2
165000 10114.5
165000 7880.5
6-3
=
=
=
=
[149161,180839
[152948,177052
[154885,175115
[157120,172880
6-13.
x = 146.75 = 35.2
95% C.I.: 146.75 1.96(35.2/ 50 ) = [136.99, 156.51]
n = 50
1.645(35.2/
99% C.I.: 146.75 2.576(35.2/
90% C.I.: 146.75
50 ) = [138.56, 154.94]
50 ) = [133.93, 159.57]

Known
Population Normal?
Population Stdev.
Sample Size
Sample Mean
(1)
99%
95%
90%
6-14.
n = 100 = 30
35.2
50
146.75
n
x-bar
Confidence Interval
146.75 12.8226
146.75 9.75675
146.75 8.18813
X = 150
90% C.I.: 150 1.645(30/ 100 ) = [145.06, 154.94]

Known
Population Normal?
Population Stdev.
Sample Size
Sample Mean
(1)
99%
95%
90%
6-15.
30
100
150
n
x-bar
Confidence Interval
150 7.7275
= [ 142.272 , 157.728 ]
150 5.87988
= [ 144.12 , 155.88 ]
150 4.93456
= [ 145.065 , 154.935 ]
n = 36 = 8%
X = 20%
95% C.I.: 20 1.96(8/ 36 ) = [17.4%, 22.6%]
6-4
=
=
=
[133.927,159.573]
[136.993,156.507]
[138.562,154.938]

Known
Population Normal?
Population Stdev.
Sample Size
Sample Mean
(1)
99%
95%
6.16.
n
x-bar
8
36
20
Confidence Interval
20 3.43445
20 2.61328
=
=
[ 16.5656 , 23.4344 ]
[ 17.3867 , 22.6133 ]
Interval width = 2 z / 2 ( / n )
Consider m = 4n. The width of the interval is now:
2 z / 2 ( / 4n ) = 2 z / 2 ( / 2 n ) = 1/2 of the old width.
Thus the answer is: 4 times the sample size.

6.17.
2(1.96 / n ) = 10
1.96( / n ) = 5
/ n = 5/1.96 = 2.551
For 90% confidence, z / 2 = 1.645

Width of 90% C.I. = 2(1.645(2.551)) = 8.393 units.
Easier way: new width = (old width)1.645/1.96 = 10(1.645)/1.96 = 8.393
6-18.
n = 50 x = 14.5 s = 5.6
95% C.I.: 14.5 1.96(5.6 /
50 ) = [12.95, 16.05]
90% C.I.: 14.5 1.645(5.6 / 50 ) = [13.20, 15.80]

Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
90%
6-19.
n = 165
Yes
50
14.5
5.6
n
x-bar
s
Confidence Interval
14.5 2.12241 = [ 12.3776 , 16.6224 ]
14.5 1.5915 = [ 12.9085 , 16.0915 ]
14.5 1.32776 = [ 13.1722 , 15.8278 ]
x = 16,530
s = 5,542
95% C.I.: 16,530 1.96(5,542 / 165 ) = [15,684.37, 17,375.63]

99% C.I.: 16,530 2.576(5,542 / 165 ) = [15,418.6, 17,641.4]
6-5
6-20.
n = 12 x = 34.2 s = 5.9
95% C.I.: x t .025(11) ( s /
n ) = 34.2 2.201(5.9 /
12 ) = [30.45, 37.95]
Assume a normal population.

6-21.
From the data we compute: n = 32
x = 30.5625 s = 5.775
99% C.I.: x z .005 ( s / n ) = 30.5625 2.576(5.775 / 32 )

= [27.93, 33.19] thousands of miles.
6-22.
n = 100
s = 2.1 X = 3.2
95% C.I.: 3.2 0.55154, [2.64546, 3.75154]
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
6-23.
n
x-bar
s
Confidence Interval
3.2 0.55154 = [ 2.64846 , 3.75154 ]
3.2 0.41669 = [ 2.78331 , 3.61669 ]
From the data compute: n = 25

For df = 24 and
Yes
100
3.2
2.1
x = 81.240
s = 15.447
= 0.01, we use t / 2 = 2.797, so
x t / 2 ( s / n ) = 81.240 2.797(15.447/5) = [72.599, 89.881]

6-24.
n = 40 x = 42,539
s = 11,690
90% C.I.: 42,539 1.645(11,690/

6.25.
40 ) = [39,498.46, 45,579.54] dollars.
Use the Estimating Mean.xls template, sheet: [Sample Data].

90% C.I.: 8.49375 3.57007, [4.92368, 12.06382]
Unknown
Population Normal?
Yes
Sample Size
16
n
Sample Stdev. 8.145958 s
(1)
99%
95%
90%
Confidence Interval
8.49375 6.00098
8.49375 4.34068
8.49375 3.57007
6-6
=
=
=
[ 2.492773 , 14.49473 ]
[ 4.153073 , 12.83443 ]
[ 4.92368 , 12.06382 ]
6-26.
n = 400
x = 212
s = 38
95% C.I.: 212 1.96(38 /
400 ) = [208.28, 215.72] calories.
98% C.I.: 212 2.326(38 /
400 ) = [207.58, 216.42] calories.
6.27.We need to assume normality of the population.

99% C.I.: 2.6 t .005(19) (0.4 /
6-28.
n = 56 x = 258
20 ) = 2.6 2.861(.4 /
20 ) = [2.344, 2.856] days.
s = 85
95% C.I.: 258 1.96(85 / 56 ) = [235.74, 280.26] dollars.

6.29.Assume a normal population.
x = 647,000
n = 200
s = 140,000
95% C.I.: 647000 19521.4, [627478.6, 666521.4]

Unknown
Population Normal?
Yes
Sample Size
200
n
Sample Mean 647000 x-bar
Sample Stdev. 140000 s
(1)
99%
95%
6-30.
n = 225
Confidence Interval
647000 25746.2 = [ 621253.8 , 672746.2 ]
647000 19521.4 = [ 627478.6 , 666521.4 ]
x = 259.6 s = 52
95% C.I.: 259.6 1.96(52 /

6-31.
225 ) = [252.81, 266.39] dollars.
n = 46 x = 16.5 s = 2.2
95% C.I.: 16.5 1.96( 2.2 /
6-32.
46 ) = [15.86, 17.14] dollars.
n = 20 x = 5,139

95% C.I.: x t .025(19) ( s /
s = 640
n ) = 5,139 2.093(640 /
= [4,839.47, 5,438.52] dollars.
6-7
20 )

Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
6.33.
Yes
20
5139
640
n
x-bar
s
Confidence Interval
5139 409.425 = [ 4729.58 , 5548.42 ]
5139 299.529 = [ 4839.47 , 5438.53 ]

n = 28 x = 6.7
s = 2.4
99% C.I.: x t .005( 27 ) ( s / n ) = 6.7 2.771( 2.4 /
6-34.
n = 75 x = 8.9
28 ) = [5.44, 7.96] years.
s = 0.5
95% C.I.: 8.9 1.96(0.5 /
75 ) = [8.79, 9.01] units.
6.35.Assume a normal distribution.

x = 61.67 s = 11.22
n = 12
90% C.I.: x t .05(11) ( s /
n ) = 61.67 1.796(11.22 /
12 ) = [55.85, 67.49] containers.

Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
90%
6-36.
Yes
12
61.67
11.22
n
x-bar
s
Confidence Interval
61.67 10.0595 = [ 51.6105 , 71.7295 ]
61.67 7.12885 = [ 54.5411 , 68.7989 ]
61.67 5.81675 = [ 55.8532 , 67.4868 ]
n = 25, s = 100 X = 600

95% C.I.: 600 41.278, [558.722, 641.278]
6-8
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
6-37.
Yes
25
600
100
n
x-bar
s
Confidence Interval
600 55.939
600 41.278
= [
= [
544.061 , 655.939
558.722 , 641.278
90% CI: [9.763642, 10.38036]

Unknown
Population Normal?
Yes
Sample Size
25
n
(1)
99%
95%
90%
6-38.
Confidence Interval
10.072 0.5041
10.072 0.37198
10.072 0.30836
n = 376
s = 12 X = 41.12
90% C.I.: [40.0996, 42.1404]
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
Yes
376
41.12
12
6-9
n
x-bar
s
=
=
=
[
[
[
9.567897 , 10.5761
9.700017 , 10.44398
9.763642 , 10.38036]
]
]
(1)
99%
95%
90%
Confidence Interval
41.12 1.60221 = [ 39.5178 , 42.7222 ]
41.12 1.21686 = [ 39.9031 , 42.3369 ]
41.12 1.02044 = [ 40.0996 , 42.1404 ]
6-10
6-39.
90% C.I.: [10.76, 12.16]
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
90%
6-40.
n = 120
Yes
376
11.46
8.2
n
x-bar
s
Confidence Interval
11.46 1.09485 = [ 10.3652 , 12.5548 ]
11.46 0.83152 = [ 10.6285 , 12.2915 ]
11.46 0.6973 = [ 10.7627 , 12.1573 ]
x = 28
95% C.I.:
1.96
p
q
/n
p
28
28 92
1.96
120
120
120 120
= [0.1577, 0.3090]
6-41.
n = 68 x = 42
99% C.I.:
2.576
p
q
/n
p
42
42 26
2.576 68
68
68 68
= [0.4658, 0.7695]
6.42.
95% CI: [0.5155, 0.5845]

(template : Estimating Proportion.xls)
Confidence Interval for Population Proportion
Sample size 800 n
Sample proportion 0.55 p-hat
1
99%
95%
6.43.
Confidence
Interval
0.55 0.0453
0.55 0.0345
=[
=[
90% CI: [0.3430, 0.4570]

(template : Estimating Proportion.xls)
6-11
0.5047,0.5953
0.5155,0.5845
]
]

Sample size 200 n
1
99%
95%
90%
6-44.
n = 280
90% C.I.:
Confidence
Interval
0.4 0.0892
0.4 0.0679
0.4 0.0570
=[
=[
=[
0.3108, 0.4892
0.3321, 0.4679
0.3430, 0.4570
]
]
]
= 1/7
1
1 6
1.645 280 = [0.1085, 0.1773]
7
7 7
6.45.95% C.I.: [0.0078, 0.0122]

(Template: Estimating Proportion.xls)
Sample size 8000 n
1
99%
95%
Confidence
Interval
0.01 0.0029
0.01 0.0022
=[
=[
6.46.90% C.I.: [0.7913, 0.9087]

Sample size 100 n
1
Confidence
Interval
6-12
0.0071, 0.0129
0.0078, 0.0122
]
]
99%
95%
90%
0.85 0.0920
0.85 0.0700
0.85 0.0587
=[
=[
=[
6-13
0.7580,0.9420
0.7800,0.9200
0.7913,0.9087
]
]
]
6.47.n = 52 x = 8
98% C.I.:
6-48.
n = 250
8
8 44
2.326 52 = [0.0375, 0.2702]
52
52 52
x = 121
99% C.I.: 0.484 2.576 (.484)(.516) / 250 = [0.4026, 0.5654]

6-49.
n = 347
x = 201
90% C.I.:
6.50.
201
1.645
347
201 146
347 = [0.5357, 0.6228]

347 347
(Template : Estimating Proportion.xls)

p = 53/250 = 0.212
90% C.I.: [0.1695, 0.2545]
Sample size 250 n
1
99%
95%
90%
6.51.
Confidence
Interval
0.212 0.0666
0.212 0.0507
0.212 0.0425
=[
=[
=[
0.1454,0.2786 ]
0.1613,0.2627 ]
0.1695,0.2545 ]

p = 130/570 = 0.2281
95% C.I.: [0.1937, 0.2625]
Sample size 570 n
1
99%
95%
Confidence
Interval
0.2281 0.0453
0.2281 0.0344
=[
=[
6-14
0.1828,0.2734 ]
0.1937,0.2625 ]
6-52.
n = 22 s 2 = 8 95% C.I. for 2 :

( n 1) s 2 (n 1) s 2
,
= [21(8)/35.48, 21(8)/10.28] = [4.74, 16.34]

2
X2
X
(Template: Estimating Variance.xls, sheet: sample stats)
Confidence Interval for Population Variance
Assumption:
The population is normally distributed
1
95% [
95% [
Sample Size 22 n
Sample Variance 8 s2
6-53.
Confidence Interval
4.73521, 16.3378]
4.73521, 16.3378]
n = 41 s 2= 102
99% C.I.: for 2 : [40(102)/66.766, 40(102)/20.707] = [61.11, 197.03]
(Template: Estimating Variance.xls, sheet: sample stats)
Assumption:
The population is normally
distributed
Sample Size 41 n
Sample Variance 102 s2
6-54.
1
99% [
Confidence Interval
61.1089, 197.039 ]
n = 60 s 2 = 1,228
Well use df = 59, approximately 60, and look in the table.
95% C.I.: for 2 : [59(1,228)/83.3, 59(1,228)/40.4817] = [869.8, 1,789.7]
6-55.
From Problem 6-21:n = 32 s = 5,775

s 2 = (5.775)2 = 33.35
99% C.I.: for 2 (using approximate df = 30):
[33.35(31)/53.7, 33.35(31)/13.8] = [19.25, 74.92]
6-56.
From Problem 6-25:n = 16 s = 8.145958

s 2 = (8.145958)2 = 66.356632
6-15
Template: Estimating Variance.xls

Assumption:
The population is normally distributed
Sample Size
16
n 1
Sample Variance 66.35663 s2 95% [
95% [
90% [
Confidence Interval
36.2098, 158.948 ]
36.2098, 158.948 ]
39.8207, 137.083 ]
The 95% C.I.: [36.2098, 158.948]

6-57.
From Problem 6-26: n = 400s = 38

Using a normal approximation to the X 2 distribution we have: X ~ N(df, 2df)
For a 95% C.I. the critical values are:
x1.2 = 1.96 2(399) = 399 1.96 2(399) = 343.63, 454.37
399(38) 2 399(38) 2
,
= [1,268.03, 1,676.68]
343.63
454.37
95% C.I. for 2 :

6-58.
From Problem 6-27:n = 20 s = 0.4

(19)(.4) 2 (19)(.4) 2
,
95% C.I. for :
= [0.0925, 0.3413]
32.8523 8.90655
2
6-59.
B = 0.05 90% confidence. Use a guess of p = 0.5

n = (1.645)2(.5)(.5)/(.05)2 = 270.6
Sample at least 271 items.
(Template: Sample Size.xls, sheet: Population mean)
Confidence Level Desired
Half-Width Desired
Population Proportion
90%
0.05
0.5
Minimum Sample Size
271
2 = 100
n = (1.96)2 100/22 = 96.04 Sample at least 97 runs.
6-60.
B=5
95% confidence
6-61.
B = 2,000 95% confidence
2 = 40,000,000
n = (1.96)2 40,000,000/(2,000)2 = 38.416
Sample at least 39 executives.
6-16

95%
Half-Width Desired 2000 B
Population Stdev. 6324.555
Minimum Sample Size
39
6-62.
B = 0.5
95% confidence
2
n = (1.96) (2)2/(.5)2 = 61.466
Sample at least 62 returns.
6-63.
An estimate of p is 0.14
B = .05
90% confidence.
2
2
n = (1.645) (.14)(.86)/(.05) = 130.322
Sample at least 131 people.
6-64.
= 50
B = 10
90% confidence
2
2
2
n = (1.645) (50) /(10) = 67.65
Therefore, sample at least 68 days.
Confidence Level
Desired
Half-Width Desired
Population Stdev.
Minimum Sample Size
6-65.
6-66.
=2
90%
10
50
68
B = 0.02 95% confidence .

Guessed p = 0.10
2
2
n = (1.96) (.10)(.90)/(.02) = 864.36
Sample at least 865 accounts.
Half-Width Desired
95%
0.02
0.1
Minimum Sample Size
865
n = 10 s = 6000
X = 32,000
95% C.I.: 32,000 2.262 (6000/ 10 ) = [27,708.16, 36,291.84]

6-67.
n = 24 s = 4.718
X = 23.5
95% C.I.: 23.5 2.069 (4.718/ 24 ) = [21.507, 25.493]
6-17
6-68.
99% CI: [$8.96M, $9.44M]

(Template: Estimating Mean.xls, sheet : sample stats)
Unknown
Population Normal?
Yes
Sample Size
200
n
Sample Mean
9.2
x-bar
(1)
99%
6-69.
Confidence Interval
9.2 0.20968 = [ 8.99032 , 9.40968 ]
P = .71
95% C.I.: .71 1.96 ( (.71)(.29) / 100 ) = [.6211, .7989]
6-70.
(Template: Estimating Means.xls, sheet: sample stats)

95% C.I.: [50.07168, 60.52832]
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
6-71.
Yes
68
55.3
21.6
n
x-bar
s
Confidence Interval
55.3 6.94455 = [ 48.35545 , 62.24455 ]
55.3 5.22832 = [ 50.07168 , 60.52832 ]

90% C.I.: [35.81417, 52.18583]
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
90%
Yes
15
44
18
n
x-bar
s
Confidence Interval
44 13.8351 = [ 30.16486 , 57.83514 ]
44 9.96808 = [ 34.03192 , 53.96808 ]
44 8.18583 = [ 35.81417 , 52.18583 ]
6-18
6-72.
P = 47 / 430 = .109
95% C.I.: .109 1.96 ( (.109)(.891) / 430 ) = [.0795, .138]
Sample size 430 n

1
99%
95%
Confidence Interval
0.109 0.0387 = [
0.109 0.0295 = [
6-73.
n = 10 s = 0.2 X = 1.2
90% C.I.: 1.2 1.833 (.2/ 10 ) = [1.084, 1.316]
6-74.
n = 500
s = 800
0.0703 ,
0.0795 ,
0.1477 ]
0.1385 ]
X = 3600
95% C.I.: 3600 1.96 (800/ 500 ) = [3529.88, 3670.12]

6-75.
P = 368 / 500 = .736

95% C.I.: .736 1.96 ( (.736)(.264) / 500 ) = [.6974, .7746]
6-76.

95% C.I.: [0.4487, 1.2173]
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
Yes
9
0.833
0.5
n
x-bar
s
Confidence Interval
0.833 0.55923 = [ 0.27377 , 1.39223 ]
0.833 0.38433 = [ 0.448666 , 1.217334 ]
6-77.
If 0.00 was in the confidence interval for 6-76, then defense attorneys could argue that the banker
had a 95% probability of making no money from his insider trading. Therefore, he did benefit
from insider information.
6-78.
P = 161 / 200 = .805

95% C.I.: .805 1.96 ( (.805)(.195) / 200 ) = [.750, .860]
6-19
6-79.
(Template: Estimating Mean.xls, Sheet: Sample Data)

95% C.I.:[7.021, 8.341]
Unknown
Population Normal?
Yes
Sample Size
10
n
(1)
99%
95%
6-80.
Confidence Interval
7.681 0.94816
7.681 0.66
=
=
[ 6.732842 , 8.629158 ]
[ 7.021004 , 8.340996 ]
95% CI: [0.4062, 0.7138]

Sample size 40 n
1
99%
95%
Confidence Interval
0.56 0.2022 = [
0.56 0.1538 = [
6-81.
n = 22 s = .25 X = .60
90% C.I.: .60 1.721 (.25/ 22 ) = [.508, .692]
6-82.
P = .80
0.3578 ,
0.4062 ,
95% C.I.: .80 1.96 ( (.80)(.20) / 200 ) = [.745, .855]

6-83.
P = .91
95% C.I.: .91 1.96 ( (.91)(.09) / 5250 ) = [.902, .918]
6-20
0.7622 ]
0.7138 ]

Sample size 5250 n

1
99%
95%
6-84.
Confidence Interval
0.91 0.0102 = [
0.91 0.0077 = [
0.8998 ,
0.9023 ,
0.9202 ]
0.9177 ]
(Template: Estimating Means.xls, Sheet: Sample Data)

90% C.I.:[169.39, 369.11]
Unknown
Population Normal?
Yes
Sample Size
8
n
(1)
99%
95%
90%
6-85.
Confidence Interval
269.25 184.448
269.25 124.633
269.25 99.8581
=
=
=
[ 84.80171 , 453.6983 ]
[ 144.6171 , 393.8829 ]
[ 169.3919 , 369.1081 ]
(Template: Estimating Proportions.xls)

95% C.I.: [ 0.5695, 0.6304]
Sample size 1000 n
1
99%
95%
6-86.
Confidence
Interval
0.6 0.0399
0.6 0.0304
=[
=[
0.5601,0.6399 ]
0.5696,0.6304 ]
(Template: Sample size.xls, sheet: Population Mean)
6-21
a) sample size = 217

Sample Size Determination for
Population Mean
Confidence Level Desired 95%

Half-Width Desired 0.002
Population Stdev. 0.015
Minimum Sample Size 217
b)
stdev
n=
st217 0.002
0.01
97
0.011 117
0.012 139
0.013 163
0.014 189
0.015 217
0.016 246
0.017 278
0.018 312
0.019 347
0.02 385
c) cost of sampling:
Popn.
Std.
Devn.
6-87.
1327
0.01
0.011
0.012
0.013
0.014
0.015
0.016
0.017
0.018
0.019
0.02
n = 20 s = 3
Cost
607
727
859
1003
1159
1327
1501
1693
1897
2107
2335
X =8
75% C.I.: 8 1.187 (3/ 20 ) = [7.76, 8.24]

6-88.
(Template: Sample Size.xls, sheet: Population Proportion)
6-22
a) sample size= 1039


Half-Width Desired 0.01 B
Population Proportion 0.04
b) Pop Prop range
Popn.
Proportion
1039
0.03
0.033
0.036
0.039
0.042
0.045
0.048
0.051
0.054
0.057
0.06
c) Pop Prop range
Popn.
Proportion
Sample
0.01
788
864
939
1015
1089
1163
1237
1310
1383
1455
1526
Cost
6259
0.03
0.033
0.036
0.039
0.042
0.045
0.048
0.051
0.054
0.057
0.06
0.01
4753
5209
5659
6115
6559
7003
7447
7885
8323
8755
9181
6-23
6-89.
(Template: Estimating Mean.xls, sheet: Sample stats)

a) 95% CI: [0.985916, 0.998084]
b) 99% CI: [0.983684, 1.000316]
Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
6-90.
Yes
20
0.992
0.013
n
x-bar
s
Confidence Interval
0.992 0.00832
0.992 0.00608
=
=
[ 0.983684 , 1.000316 ]
[ 0.985916 , 0.998084 ]
95% CI: [0.0268, 0.1332]

Sample size 100 n
1
99%
95%
6-91
( n ) = 3.74 then ( n ) = 3.74 z

n ) = 1.96( 2.27356 ) = 4.456
Since z 90
z 95
Confidence Interval
0.08 0.0699 = [
0.08 0.0532 = [
90
= 3.74
The 95% C.I. :172.58 4.456
6-24
1.645
0.0101 ,
0.0268 ,
= 2.27356
0.1499 ]
0.1332 ]
6-92
(Template found in Sample Size.xls, sheet: Population Proportion)
Sample Size Determination for Population Mean
a. minimum sample sizes
Tabulatio
n of
43
14
15
16
17
18
19
20
21
22
23
24
Popn. Std.
Devn.
2
189
217
246
278
312
347
385
424
465
509
554
3
84
97
110
124
139
155
171
189
207
226
246
Half Width Desired

4
5
48
31
55
35
62
40
70
45
78
50
87
56
97
62
106
68
117
75
128
82
139
89
6
21
25
28
31
35
39
43
48
52
57
62
7
16
18
21
23
26
29
32
35
38
42
46
6
434
450
462
474
490
506
522
542
558
578
598
7
414
422
434
442
454
466
478
490
502
518
534
b. Sampling costs:
Cost Analysis
Sampling Cost $
522
Tabulatio
n of
Popn. Std.
Devn.
522
14
15
16
17
18
19
20
21
22
23
24
2
1106
1218
1334
1462
1598
1738
1890
2046
2210
2386
2566
3
686
738
790
846
906
970
1034
1106
1178
1254
1334
Half Width
4
542
570
598
630
662
698
738
774
818
862
906
Desired
5
474
490
510
530
550
574
598
622
650
678
706
6-25
c. Total costs:
Cost Analysis
Sampling Cost $
522
Error Cost $144,000
Total Cost $144,522
Tabulatio
n of
Popn. Std.
Devn.
144522
14
15
16
17
18
19
20
21
22
23
24
2
1146
1258
1374
1502
1638
1778
1930
2086
2250
2426
2606
3
776
828
880
936
996
1060
1124
1196
1268
1344
1424
Half Width
4
702
730
758
790
822
858
898
934
978
1022
1066
Desired
5
724
740
760
780
800
824
848
872
900
928
956
6
794
810
822
834
850
866
882
902
918
938
958
The value of B that minimizes the total costs when =14 is 4

The value of B that minimizes the total costs when =24 is 5
6-93
(Template: Sample Size.xls, sheet: Population Proportion)
Sample Size Determination for Population Proportion
minimum sample size = 1624
Half-Width Desired
90%
0.02
0.6
Minimum Sample Size
1624
6-26
7
904
912
924
932
944
956
968
980
992
1008
1024
6.94.

95% C.I.: [0.4862, 0.5338]
Sample size 1700 n
1
99%
95%
6.95.
Confidence
Interval
0.51 0.0312
0.51 0.0238
=[
=[
0.4788,0.5412 ]
0.4862,0.5338 ]
95% C.I.: [5.147, 6.853]

Unknown
Population Normal?
Sample Size
Sample Mean
Sample Stdev.
(1)
99%
95%
Yes
50
6
3
n
x-bar
s
Confidence Interval
6 1.13701 = [ 4.86299 , 7.13701 ]
6 0.85259 = [ 5.14741 , 6.85259 ]
Case 7: Presidential Polling

1) sample size = 2393

Half-Width Desired 0.02 B
Population Proportion 0.53
6-27
2)
Tabulatio
n of
Popn.
Proportion
2393
0.4
0.42
0.44
0.46
0.48
0.5
0.52
0.54
0.56
0.58
0.6
0.01
9220
9358
9466
9543
9589
9604
9589
9543
9466
9358
9220
0.014
4704
4775
4830
4869
4892
4900
4892
4869
4830
4775
4704
Half Width
0.018
2846
2889
2922
2946
2960
2965
2960
2946
2922
2889
2846
Desired
0.022
1905
1934
1956
1972
1982
1985
1982
1972
1956
1934
1905
0.026
1364
1385
1401
1412
1419
1421
1419
1412
1401
1385
1364
0.03
1025
1040
1052
1061
1066
1068
1066
1061
1052
1040
1025
3) As the half-width increases, the sample size decreases.

As the population proportion approaches 50% (from either direction), the sample size increases.
4) 50%
5) cost = $425 + $1.20(n)
Tabulatio
n of
Popn.
Proportion
3296.6
0.4
0.42
0.44
0.46
0.48
0.5
0.52
0.54
0.56
0.58
0.6
0.01
11489
11654.6
11784.2
11876.6
11931.8
11949.8
11931.8
11876.6
11784.2
11654.6
11489
0.014
6069.8
6155
6221
6267.8
6295.4
6305
6295.4
6267.8
6221
6155
6069.8
Half Width Desired

0.018
0.022
3840.2
2711
3891.8 2745.8
3931.4 2772.2
3960.2 2791.4
3977
2803.4
3983
2807
3977
2803.4
3960.2 2791.4
3931.4 2772.2
3891.8 2745.8
3840.2
2711
6-28
0.026
2061.8
2087
2106.2
2119.4
2127.8
2130.2
2127.8
2119.4
2106.2
2087
2061.8
0.03
1655
1673
1687.4
1698.2
1704.2
1706.6
1704.2
1698.2
1687.4
1673
1655
6) Cost for reducing the error to 1% will increase sampling costs by a factor of approximately 3.60 from
the current costs associated to a 2% error.
Tabulation
of
Popn.
Proportion
3296.6
0.4
0.42
0.44
0.46
0.48
0.5
0.52
0.54
0.56
0.58
0.6
0.01
11489
11654.6
11784.2
11876.6
11931.8
11949.8
11931.8
11876.6
11784.2
11654.6
11489
0.012
8108.6
8223.8
8313.8
8377.4
8415.8
8429
8415.8
8377.4
8313.8
8223.8
8108.6
Half Width Desired

0.014
0.016
6069.8 4747.4
6155
4812.2
6221
4862.6
6267.8 4898.6
6295.4 4920.2
6305
4927.4
6295.4 4920.2
6267.8 4898.6
6221
4862.6
6155
4812.2
6069.8 4747.4
0.018
3840.2
3891.8
3931.4
3960.2
3977
3983
3977
3960.2
3931.4
3891.8
3840.2
factor
0.02 increase
3191
3.600439
3233
3.604887
3265.4 3.608807
3288.2 3.611885
3302.6
3.61285
3306.2 3.614361
3302.6
3.61285
3288.2 3.611885
3265.4 3.608807
3233
3.604887
3191
3.600439
Case 8: Privacy Problem

(Using the template: Sample Size.xls, sheet: Population Proportion)
[Note: this case is dealing with the number of attempts to extract private information and not the number
of calls being placed. So the sample size reflects the number of attempts being made.]
1. minimum sample size (no. of attempts)
is 626, which is not evenly divisible by 3
Half-Width Desired
95%
0.02
0.07
Minimum Sample Size
626
Total Cost = 380+0.75(626)= $850
B
Cost Analysis
Sampling Cost $ 850
Error Cost
Total Cost $ 850
2. minimum sample size (no. of attempts)

is 2501, which is also not evenly divisible by 3
Half-Width Desired
95%
0.01
0.07
Minimum Sample Size
2501
Total Cost =380+0.45(2501)= $2256
6-29
Cost Analysis
Sampling Cost $ 2256
Error Cost
Total Cost $ 2256
6-30
3. Tabulation of Total Cost

Tabulatio
n of
Popn.
Proportion
2255.75
0.05
0.055
0.06
0.065
0.07
0.075
0.08
0.085
0.09
0.095
0.1
0.01
1748.75
1877.75
2005.25
2131.25
2255.75
2379.5
2501
2621
2740.25
2857.25
2973.5
Half Width Desired

0.014
0.018
0.022
1078.25
803
663.5
1144.25 842.75 689.75
1209.5 881.75
716
1274
920.75 742.25
1337
959
767.75
1400
997.25 793.25
1462.25 1034.75 818.75
1523.75 1072.25 843.5
1584.5
1109
868.25
1644.5
1145
892.25
1703
1181
916.25
0.026
582.5
602
620.75
639.5
657.5
676.25
694.25
711.5
729.5
746.75
764
0.03
532.25
546.5
560.75
575
588.5
602.75
616.25
629
642.5
655.25
668.75
Graph
3000
2500
2000
Total Cost
1500
1000
0.05
0.06
500
0.07
0.08
0
0.09
0.01
0.014
0.018
0.022
Popn.
Proportion
0.1
0.026
Half Width
0.03
[Note: in parts 4) and 5), the cost per attempt drops to $0.45. Therefore, the Total Cost will drop.]
4. Although the total number of attempts is not evenly divisible by 5, the Total Cost drops to $662
Cost Analysis
Sampling Cost $
Error Cost
Total Cost $
662
662
6-31
5. Although the total number of attempts is not evenly divisible by 5, the Total Cost drops to $1505
Cost Analysis
Sampling Cost $ 1,505
Error Cost
Total Cost $ 1,505
6. Problems: One may be successful in attaining private information; the call could last longer driving the
cost per call above $2.25; the true proportion could increase above 7%.
6-32

Chap 006

Uploaded by

Copyright:

Available Formats

Chap 006

Uploaded by

Document Information

Original Description:

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

Chap 006

Uploaded by

Copyright:

Available Formats

Chapter 06 - Confidence Intervals

A confidence interval is an interval of values believed to contain an unknown population

= 89,673.12 2,695 = [86,978.12, 92,368.12] dollars.

Chapter 06 - Confidence Intervals

.9016 = [31.098, 32.902] m.p.g.

= [9.045, 9.555] percent alcohol.

(template : Estimating Mean.xls, sheet : sample stats)

6-10. = 430 n = 30 x = 1482

Chapter 06 - Confidence Intervals

The 90% C.I.: 1700 186.094 [1513.91, 1886.09]

80% C.I. for : 165000 7880.5, [157120, 172880]

Chapter 06 - Confidence Intervals

(template : Estimating Mean.xls, sheet : sample stats)

90% C.I.: 150 1.645(30/ 100 ) = [145.06, 154.94]

95% C.I.: 20 1.96(8/ 36 ) = [17.4%, 22.6%]

Chapter 06 - Confidence Intervals

(template : Estimating Mean.xls, sheet : sample stats)

Thus the answer is: 4 times the sample size.

For 90% confidence, z / 2 = 1.645

90% C.I.: 14.5 1.645(5.6 / 50 ) = [13.20, 15.80]

95% C.I.: 16,530 1.96(5,542 / 165 ) = [15,684.37, 17,375.63]

Chapter 06 - Confidence Intervals

Assume a normal population.

From the data we compute: n = 32

99% C.I.: x z .005 ( s / n ) = 30.5625 2.576(5.775 / 32 )

From the data compute: n = 25

= 0.01, we use t / 2 = 2.797, so

x t / 2 ( s / n ) = 81.240 2.797(15.447/5) = [72.599, 89.881]

90% C.I.: 42,539 1.645(11,690/

40 ) = [39,498.46, 45,579.54] dollars.

Use the Estimating Mean.xls template, sheet: [Sample Data].

Chapter 06 - Confidence Intervals

95% C.I.: 212 1.96(38 /

400 ) = [208.28, 215.72] calories.

98% C.I.: 212 2.326(38 /

400 ) = [207.58, 216.42] calories.

6.27.We need to assume normality of the population.

20 ) = [2.344, 2.856] days.

95% C.I.: 258 1.96(85 / 56 ) = [235.74, 280.26] dollars.

95% C.I.: 647000 19521.4, [627478.6, 666521.4]

95% C.I.: 259.6 1.96(52 /

225 ) = [252.81, 266.39] dollars.

46 ) = [15.86, 17.14] dollars.

Assume a normal population.

= [4,839.47, 5,438.52] dollars.

Chapter 06 - Confidence Intervals

(template : Estimating Mean.xls, sheet : sample stats)

Assume a normal population.

28 ) = [5.44, 7.96] years.

95% C.I.: 8.9 1.96(0.5 /

75 ) = [8.79, 9.01] units.

6.35.Assume a normal distribution.

12 ) = [55.85, 67.49] containers.

(template : Estimating Mean.xls, sheet : sample stats)

n = 25, s = 100 X = 600

Chapter 06 - Confidence Intervals

90% CI: [9.763642, 10.38036]

Chapter 06 - Confidence Intervals

Chapter 06 - Confidence Intervals