Professional Documents
Culture Documents
Chapter 3 Descriptive Statistics: Numerical Methods
Chapter 3 Descriptive Statistics: Numerical Methods
Descriptive Statistics: Numerical Methods
Learning Objectives
1. Understand the purpose of measures of location.
2. Be able to compute the mean, median, mode, quartiles, and various percentiles.
3. Understand the purpose of measures of variability.
4. Be able to compute the range, interquartile range, variance, standard deviation, and coefficient of
variation.
5. Understand skewness as a measure of the shape of a data distribution. Learn how to recognize when
a data distribution is negatively skewed, roughly symmetric, and positively skewed.
6. Understand how z scores are computed and how they are used as a measure of relative location of a
data value.
7. Know how Chebyshev’s theorem and the empirical rule can be used to determine the percentage of
the data within a specified number of standard deviations from the mean.
8. Learn how to construct a 5number summary and a box plot.
9. Be able to compute and interpret covariance and correlation as measures of association between two
variables.
10. Be able to compute a weighted mean.
3 1
Chapter 3
Solutions:
xi 75
1. x 15
n 5
10, 12, 16, 17, 20
Median = 16 (middle value)
xi 96
2. x 16
n 6
10, 12, 16, 17, 20, 21
16 17
Median = 16.5
2
3. 15, 20, 25, 25, 27, 28, 30, 34
20
i (8) 1.6 2nd position = 20
100
25 20 25
i (8) 2 22.5
100 2
65
i (8) 5.2 6th position = 28
100
75 28 30
i (8) 6 29
100 2
xi 657
4. Mean 59.727
n 11
Median = 57 6th item
Mode = 53 It appears 3 times
3 2
Descriptive Statistics: Numerical Methods
5. a. x
x i
1194 .50
39.82
n 30
b. There are an even number of items. Thus, the median is the average of the 15th and 16th items after
the data have been placed in rank order.
40.2 40.8
Median = 2 = 40.5
c. Mode = 42.0 This value appears 3 times
25
d. First Quartile i 30 7.5
100
Rounding up, we see that Q1 is at the 8th position.
Q1 = 38.2
75
e. Third Quartile i 30 22.5
100
Rounding up, we see that Q3 is at the 23rd position.
Q3 = 42.1
xi 871.74
6. a. x 36.32
n 24
Median is average of 10th and 11th values after arranging in ascending order.
39.00 39.95
Median 39.48
2
Data are multimodal
xi 491.14
b. x 20.46
n 24
19.75 19.95
Median 19.85
2
Data are bimodal: 19.95 (3 brokers), 29.95 (3 brokers)
c. Comparing the measures of central location, we conclude that it costs more to trade 100 shares in a
broker assisted trade than 500 shares online.
3 3
Chapter 3
d. From the data we have here it is more related to whether the trade is brokerassisted or online. The
amount of the online transaction is 5 times as great but the cost of the transaction is less.
However, if the comparison was restricted to brokerassisted or online trades, we would probably
find that larger transactions cost more.
xi 1380
7. a. x 46
n 30
b. Yes, the mean here is 46 minutes. The newspaper reported an average of 45 minutes.
45 52.9
c. Median 48.95
2
d. Q1 = 7 (value of 8th item in ranked order)
Q3 = 70.4 (value of 23rd item in ranked list)
40
e. Find position i 30 12; 40th percentile is average of values in 12th and 13th positions.
100
8. a. xi = 695
xi 695
x 34.75
n 20
The modal age is 29; it appears 3 times.
b. Median is average of 10th and 11th items.
33 36
Median 34.5
2
Data suggest at home workers are slightly younger.
c. For Q1,
25
i 20 5
100
Since i is integer,
3 4
Descriptive Statistics: Numerical Methods
25 26
Q1 25.5
2
For Q3,
75
i 20 15
100
Since i is integer,
42 45
Q3 43.5
2
32
d. i 20 6.4
100
Since i is not an integer, we round up to the 7th position.
32nd percentile = 27
xi 270,377
9. a. x 10,815.08 Median (Position 13) = 8296
n 25
b. Median would be better because of large data values.
c. i = (25 / 100) 25 = 6.25
Q1 (Position 7) = 5984
i = (75 / 100) 25 = 18.75
Q3 (Position 19) = 14,330
d. i = (85/100) 25 = 21.25
85th percentile (position 22) = 15,593. Approximately 85% of the websites have less than 15,593
unique visitors.
10. a. xi = 435
xi 435
x 48.33
n 9
3 5
Chapter 3
Data in ascending order:
28 42 45 48 49 50 55 58 60
Median = 49
Do not report a mode; each data value occurs once.
The index could be considered good since both the mean and median are less than 50.
25
b. i 9 2.25
100
Q1 (3rd position) = 45
75
i 9 6.75
100
Q3 (7th position) = 55
For the samples we see that the mean mileage is better in the country than in the city.
City
13.2 14.4 15.2 15.3 15.3 15.3 15.9 16 16.1 16.2 16.2 16.7 16.8
Median
Mode: 15.3
Country
17.2 17.4 18.3 18.5 18.6 18.6 18.7 19.0 19.2 19.4 19.4 20.6 21.1
Median
Mode: 18.6, 19.4
The median and modal mileages are also better in the country than in the city.
3 6
Descriptive Statistics: Numerical Methods
xi 1976
b. x 98.8 pictures
n 20
xi 2204
c. x 110.2 minutes
n 20
d. This is not an easy choice because it is a multicriteria problem. If price was the only criterion, the
lowest price camera (Fujifilm DX10) would be preferred. If maximum picture capacity was the
only criterion, the maximum picture capacity camera (Kodak DC280 Zoom) would be preferred.
But, if battery life was the only criterion, the maximum battery life camera (Fujifilm DX10) would
be preferred. There are many approaches used to select the best choice in a multicriteria situation.
These approaches are discussed in more specialized books on decision analysis.
13. Range 20 10 = 10
10, 12, 16, 17, 20
25
i (5) 1.25
100
Q1 (2nd position) = 12
75
i (5) 3.75
100
Q3 (4th position) = 17
IQR = Q3 Q1 = 17 12 = 5
xi 75
14. x 15
n 5
( xi x ) 2 64
s2 16
n 1 4
s 16 4
3 7
Chapter 3
25 20 25
i (8) 2 Q1 22.5
100 2
75 28 30
i (8) 6 Q3 29
100 2
IQR = Q3 Q1 = 29 22.5 = 6.5
xi 204
x 25.5
n 8
( xi x ) 2 242
s2 34.57
n 1 7
s 34.57 588
.
16. a. Range = 190 168 = 22
b. ( xi x ) 2 376
s 2 = 376 = 75.2
5
c. s 75.2 8.67
8.67
d. Coefficient of Variation 100% 4.87%
178
17. Range = 9267 = 25
IQR = Q3 Q1 = 80 77 = 3
x = 78.4667
x x
2
i 411.7333
x x
2
i 411.7333
s 2
29.4095
n 1 14
s 29.4095 5.4231
x
18. a. x i 115.13 (Mainland); 36.62 (Asia)
n
Median (7th and 8th position)
Mainland = (110.87 + 112.25) / 2 = 111.56
3 8
Descriptive Statistics: Numerical Methods
Median (6th and 7th position)
Asia = (32.98 + 40.41) / 2 = 36.695
b. Range = High Low
Mainland Asia
Range 86.24 42.97
Standard Deviation 26.82 11.40
Coefficient of Variation 23.30% 31.13%
c. Greater mean and standard deviation for Mainland. Greater coefficient of variation for Asia.
19. a. Range = 60 28 = 32
IQR = Q3 Q1 = 55 45 = 10
435
b. x 48.33
9
( xi x ) 2 742
( xi x ) 2 742
s2 92.75
n 1 8
s 92.75 9.63
c. The average air quality is about the same. But, the variability is greater in Anaheim.
20. Dawson Supply: Range = 11 9 = 2
4.1
s 0.67
9
J.C. Clark: Range = 15 7 = 8
60.1
s 2.58
9
3 9
Chapter 3
21. a. Winter
Range = 21 - 12 = 9
IQR = Q3 - Q1 = 20-16 = 4
Summer
Range = 38 18 = 20
IQR = Q3 - Q1 = 29-18 = 11
b.
Variance Standard Deviation
Winter 8.2333 2.8694
Summer 44.4889 6.6700
c. Winter
s 2.8694
Coefficient of Variation = 100% 100% 16.21%
x
17.7
Summer
s 6.6700
Coefficient of Variation = 100% 100% 26.05%
x
25.6
d. More variability in the summer months.
22. a. 100 Shares at $50 (Brokerassisted)
Min Value = 9.95 Max Value = 55.00
Range = 55.00 9.95 = 45.05
Interquartile range = 48.975 24.995 = 23.98
500 Shares at $50 (Online)
Min Value = 5.00 Max Value = 62.50
Range = 62.50 5.00 = 57.50
Interquartile range = 24.950 13.475 = 11.475
b. 100 Shares at $50 (Brokerassisted)
3 10
Descriptive Statistics: Numerical Methods
( xi x ) 2
s2 190.67
n 1
s s 2 13.81
500 Shares at $50 (Online)
( xi x ) 2
s2 140.63
n 1
s s 2 11.86
c. 100 Shares at $50 (Brokerassisted)
s 13.81
Coefficient of Variation = (100%) (100%) 38.02%
x 36.32
500 Shares at $50 (Online)
s 11.86
Coefficient of Variation = (100%) (100%) 57.97%
x 20.46
d. Using the standard deviation as a measure, the variability seems to be greater for the brokerassisted
trades. But, using the coefficient of variation as a measure, we see that the relative variability is
greater for the online trades.
23. s2 = 0.0021 Production should not be shut down since the variance is less than .005.
24. Quarter milers
s = 0.0564
Coefficient of Variation = (s/ x )100% = (0.0564/0.966)100% = 5.8%
Milers
s = 0.1295
Coefficient of Variation = (s/ x )100% = (0.1295/4.534)100% = 2.9%
Yes; the coefficient of variation shows that as a percentage of the mean the quarter milers’ times
show more variability.
xi 75
25. x 15
n 5
3 11
Chapter 3
( xi x ) 2 64
s2 4
n 1 4
10 15
10 z 1.25
4
20 15
20 z 1.25
4
12 15
12 z 0.75
4
17 15
17 z .50
4
16 15
16 z .25
4
520 500
26. z .20
100
650 500
z 1.50
100
500 500
z 0.00
100
450 500
z 0.50
100
280 500
z 2.20
100
40 30 1
27. a. z 2 1 0.75 At least 75%
5 22
45 30 1
b. z 3 1 0.89 At least 89%
5 32
38 30 1
c. z 1.6 1 0.61 At least 61%
5 1.6 2
42 30 1
d. z 2.4 1 0.83 At least 83%
5 2.4 2
48 30 1
e. z 3.6 1 0.92 At least 92%
5 3.6 2
3 12
Descriptive Statistics: Numerical Methods
28. a. Approximately 95%
b. Almost all
c. Approximately 68%
29. a. This is from 2 standard deviations below the mean to 2 standard deviations above the mean.
1 1 1 3
1 2
1 2 1
z 2 4 4
Therefore, at least 75% of adults sleep between 6.8 and 10.0 hours per day.
b. This is from 2.5 standard deviations below the mean to 2.5 standard deviations above the mean.
1 1 1
1 2
1 2
1 .84
z 2.5 6.25
Therefore, at least 84% of adults sleep between 6.4 and 10.4 hours per day.
c. With z = 2, the empirical rule suggests that 95% of adults sleep between 6.8and 10.0 hours per day.
The percentage obtained using the empirical rule is greater than the percentage obtained using
Chebyshev’s theorem.
30. a. 2 hours is 1 standard deviation below the mean. Thus, the empirical rule suggests that 68% of the
kids watch television between 2 and 4 hours per day. Since a bellshaped distribution is symmetric,
approximately, 34% of the kids watch television between 2 and 3 hours per day.
b. 1 hour is 2 standard deviations below the mean. Thus, the empirical rule suggests that 95% of the
kids watch television between 1 and 5 hours per day. Since a bellshaped distribution is symmetric,
approximately, 47.5% of the kids watch television between 1 and 3 hours per day. In part (a) we
concluded that approximately 34% of the kids watch television between 2 and 3 hours per day; thus,
approximately 34% of the kids watch television between 3 and 4 hours per day. Hence,
approximately 47.5% + 34% = 81.5% of kids watch television between 1 and 4 hours per day.
c. Since 34% of the kids watch television between 3 and 4 hours per day, 50% 34% = 16% of the
kids watch television more than 4 hours per day.
31. a. Approximately 68% of scores are within 1 standard deviation from the mean.
b. Approximately 95% of scores are within 2 standard deviations from the mean.
c. Approximately (100% 95%) / 2 = 2.5% of scores are over 130.
d. Yes, almost all IQ scores are less than 145.
3 13
Chapter 3
71.00 90.06
32. a. z 0.95
20
168 90.06
b. z 3.90
20
c. The zscore in part a indicates that the value is 0.95 standard deviations below the mean. The z
score in part b indicates that the value is 3.90 standard deviations above the mean.
The labor cost in part b is an outlier and should be reviewed for accuracy.
b. This is from 2 standard deviations below the mean to 2 standard deviations above the mean.
With z = 2, Chebyshev’s theorem gives:
1 1 1 3
1 1 2 1
z2 2 4 4
Therefore, at least 75% of benefits managers have an annual salary between NT$36,580 and
NT$50,380.
c. The histogram of the salary data is shown below:
3 14
Descriptive Statistics: Numerical Methods
Visual inspection of the histogram and the skewness measure of .53 indicate that it is moderately
skewed to the right. Although the distribution is not perfectly bell shaped, it does appear that the
distribution of annual salaries for benefit managers is NOT symmetric.
e. There are no outliers because all the observations are within 3 standard deviations of the mean.
b. If the distribution is bell shaped with a mean of 100 points, the percentage of NBA games in which
the winning team scores more than 100 points is 50%. A score of 114 points is z = 1 standard
deviation above the mean. Thus, the empirical rule suggests that 68% of the winning teams will
score between 86 and 114 points. In other words, 32% of the winning teams will score less than 86
points or more than 114 points. Because a bellshaped distribution is symmetric, approximately
16% of the winning teams will score more than 114 points.
c. For the winning margin, x is 11.1 and s is 10.77. To see if there are any outliers, we will first
compute the zscore for the winning margin that is farthest from the sample mean of 11.1, a winning
margin of 32 points.
x x 32 11.1
z 1.94
s 10.77
Thus, a winning margin of 32 points is not an outlier (z = 1.94 < 3). Because a winning margin of
32 points is farthest from the mean, none of the other data values can have a zscore that is less than
3 or greater than 3 and hence we conclude that there are no outliers
xi 79.86
35. a. x 3.99
n 20
4.17 4.20
Median = 4.185 (average of 10th and 11th values)
2
b. Q1 = 4.00 (average of 5th and 6th values)
Q3 = 4.50 (average of 15th and 16th values)
( xi x ) 2 12.5080
c. s 0.8114
n 1 19
3 15
Chapter 3
d. The distribution is significantly skewed to the left.
4.12 3.99
e. Allison One: z 0.16
0.8114
2.32 3.99
Omni Audio SA 12.3: z 2.06
0.8114
f. The lowest rating is for the Bose 501 Series. It’s zscore is:
2.14 3.99
z 2.28
0.8114
This is not an outlier so there are no outliers.
36. 15, 20, 25, 25, 27, 28, 30, 34
Smallest = 15
25 20 25
i (8) 2 Q1 22.5
100 2
25 27
Median 26
2
75 28 30
i (8) 8 Q3 29
100 2
Largest = 34
37.
15 20 25 30 35
38. 5, 6, 8, 10, 10, 12, 15, 16, 18
Smallest = 5
25
i (9) 2.25 Q1 = 8 (3rd position)
100
3 16
Descriptive Statistics: Numerical Methods
Median = 10
75
i (9) 6.75 Q3 = 15 (7th position)
100
Largest = 18
5 10 15 20
39. IQR = 50 42 = 8
Lower Limit: Q1 1.5 IQR = 42 12 = 30
Upper Limit: Q3 + 1.5 IQR = 50 + 12 = 62
65 is an outlier
40. a. Five number summary: 5 9.6 14.5 19.2 52.7
b. IQR = Q3 Q1 = 19.2 9.6 = 9.6
Lower Limit: Q1 1.5 (IQR) = 9.6 1.5(9.6) = 4.8
Upper Limit: Q3 + 1.5(IQR) = 19.2 + 1.5(9.6) = 33.6
c. The data value 41.6 is an outlier (larger than the upper limit) and so is the data value 52.7. The
financial analyst should first verify that these values are correct. Perhaps a typing error has caused
25.7 to be typed as 52.7 (or 14.6 to be typed as 41.6). If the outliers are correct, the analyst might
consider these companies with an unusually large return on equity as good investment candidates.
d.
* *
10 5 20 35 50 65
41. a. Median (11th position) 4019
3 17
Chapter 3
25
i (21) 5.25
100
Q1 (6th position) = 1872
75
i (21) 15.75
100
Q3 (16th position) = 8305
608, 1872, 4019, 8305, 14138
b. Limits:
IQR = Q3 Q1 = 8305 1872 = 6433
Lower Limit: Q1 1.5 (IQR) = 7777
Upper Limit: Q3 + 1.5 (IQR) = 17955
c. There are no outliers, all data are within the limits.
d. Yes, if the first two digits in Johnson and Johnson's sales were transposed to 41,138, sales would
have shown up as an outlier. A review of the data would have enabled the correction of the data.
e.
42. a. Mean = 105.7933 Median = 52.7
b. Q1 = 15.7 Q3 = 78.3
c. IQR = Q3 Q1 = 78.3 15.7 = 62.6
Lower limit for box plot = Q1 1.5(IQR) = 15.7 1.5(62.6) = 78.2
Upper limit for box plot = Q3 + 1.5 (IQR) = 78.3 + 1.5(62.6) = 172.2
Note: Because the number of shares covered by options grants cannot be negative, the lower limit
for the box plot is set at 0. Thus, outliers are values in the data set greater than 172.2.
3 18
Descriptive Statistics: Numerical Methods
Outliers: Silicon Graphics (188.8) and ToysRUs (247.6)
d. Mean percentage = 26.73. The current percentage is much greater.
43. a. Five Number Summary (Midsize)
51 71.5 81.5 96.5 128
Five Number Summary (Small)
73 101 108.5 121 140
b. Box Plots
Midsize
Small Size
c. The midsize cars appear to be safer than the small cars.
44. a. x = 37.48 Median = 23.67
b. Q1 = 7.91 Q3 = 51.92
c. IQR = 51.92 7.91 = 44.01
Lower Limit: Q1 1.5(IQR) = 7.91 1.5(44.01) = 58.11
Upper Limit: Q3 + 1.5(IQR) = 51.92 + 1.5(44.01) = 117.94
Russia, with a percent change of 125.89, is an outlier.
Turkey, with a percent change of 254.45 is another outlier.
d. With a percent change of 22.64, the United States is just below the 50th percentile the median.
3 19
Chapter 3
70
60
50
40
y
30
20
10
0
0 5 10 15 20
x
45. a.
b. Negative relationship
40 230
c/d. xi 40 x 8 yi 230 y 46
5 5
( xi x )( yi y ) 240
sxy 60
n 1 5 1
( xi x ) 2 118
sx 5.4314
n 1 5 1
( yi y ) 2 520
sy 11.4018
n 1 5 1
sxy 60
rxy 0.969
sx s y (5.4314)(11.4018)
There is a strong negative linear relationship.
3 20
Descriptive Statistics: Numerical Methods
46. a.
18
16
14
12
10
y
8
6
4
2
0
0 5 10 15 20 25 30
x
b. Positive relationship
80 50
c/d. xi 80 x 16 yi 50 y 10
5 5
( xi x )( yi y ) 106 ( xi x ) 2 272 ( yi y ) 2 86
( xi x )( yi y ) 106
sxy 26.5
n 1 5 1
( xi x ) 2 272
sx 8.2462
n 1 5 1
( yi y ) 2 86
sy 4.6368
n 1 5 1
sxy 26.5
rxy 0.693
sx s y (8.2462)(4.6368)
A positive linear relationship
3 21
Chapter 3
47. a.
750
700
650
y = SAT
600
550
500
450
400
2.6 2.8 3 3.2 3.4 3.6 3.8
x = GPA
b. Positive relationship
19.8 3540
c/d. xi 19.8 x 3.3 yi 3540 y 590
6 6
( xi x )( yi y ) 143
sxy 28.6
n 1 6 1
( xi x ) 2 0.74
sx 0.3847
n 1 6 1
( yi y ) 2 36, 400
sy 85.3229
n 1 6 1
sxy 28.6
rxy 0.8713
sx s y (0.3847)(85.3229)
A positive linear relationship
48. Let x = driving speed and y = mileage
3 22
Descriptive Statistics: Numerical Methods
420 270
xi 420 x 42 yi 270 y 27
10 10
( xi x ) 2 1660
sx 13.5810
n 1 10 1
( yi y ) 2 164
sy 4.2687
n 1 10 1
s xy 52.7778
rxy .91
sx s y (13.5810)(4.2687)
A strong negative linear relationship
49. a. The sample correlation coefficient is .78.
b. There is a positive linear relationship between the performance score and the overall rating.
51. The sample correlation coefficient is .88. This indicates a strong positive linear relationship
between the daily high and low temperatures.
53.
fi Mi fi Mi
4 5 20
7 10 70
9 15 135
5 20 100
25 325
f i M i 325
x 13
n 25
3 23
Chapter 3
fi Mi Mi x (M i x )2 fi ( M i x )2
4 5 8 64 256
7 10 3 9 63
9 15 +2 4 36
5 20 +7 49 245
600
f i ( M i x ) 2 600
s2 25
n 1 24
s 25 5
54. a.
Grade xi Weight Wi
4 (A) 9
3 (B) 15
2 (C) 33
1 (D) 3
0 (F) 0
60 Credit Hours
b. Yes; satisfies the 2.5 grade point average requirement
440,500
17.62%
25, 000
31, 298
1.25%
25, 000
56.
Mi fi fi Mi Mi x (M i x )2 fi ( M i x ) 2
2 74 148 -8.742647 76.433877 5,656.1069
7 192 1,344 -3.742647 14.007407 2,689.4221
12 280 3,360 1.257353 1.580937 442.6622
17 105 1,785 6.257353 39.154467 4,111.2190
22 23 506 11.257353 126.728000 2,914.7439
27 6 162 16.257353 264.301530 1,585.8092
680 7,305 17,399.9630
3 24
Descriptive Statistics: Numerical Methods
7305
x 10.74
680
17,399.9630
s2 25.63
679
s 5.06
Estimate of total gallons sold: (10.74)(120) = 1288.8
57. a.
Class fi Mi fi Mi
0 15 0 0
1 10 1 10
2 40 2 80
3 85 3 255
4 350 4 1400
Totals 500 1745
i fM i 1745
x 3.49
n 500
b.
Mi x ( Mi x ) 2 fi ( Mi x ) 2
3.49 12.18 182.70
2.49 6.20 62.00
1.49 2.22 88.80
0.49 0.24 20.41
+0.51 0.26 91.04
Total 444.95
( M i x ) 2 f i 444.95
s2 0.8917 s 0.8917 0.9443
n 1 499
58. a. x
x i
37640
1505.6
n 25
Median = 1620 (13th value)
Mode = 0 (4 times)
b. It appears that this group of households spent much less than the average household. The mean and
median are much lower than the average of NT$ 2,117 reported in the newspaper.
c. Q1 = 982 (7th value)
3 25
Chapter 3
Q3 = 2100 (19th value)
d. Min = 0 Max = 3280
Range = 3280 0 = 3280
IQR = Q3 Q1 = 2100 982 = 1118
e. s2 = 847722.3 s = 920.72
f. The data are significantly skewed to the left. This seems reasonable. A few people buy clothes and
footwear almost all the time causing the left tail of the distribution to be longer, and the distribution
is bounded below at zero.
g. The z score for the largest value is:
3280 1505.6
z 1.93
920.72
It is the only outlier and should be checked for accuracy.
59. a. xi = 760
xi 760
x 38
n 20
Median is average of 10th and 11th items.
36 36
Median 36
2
The modal cash retainer is 40; it appears 4 times.
b. For Q1,
25
i 20 5
100
Since i is integer,
28 30
Q1 29
2
For Q3,
3 26
Descriptive Statistics: Numerical Methods
75
i 20 15
100
Since i is integer,
40 50
Q3 45
2
c. Range = 64 – 15 = 49
Interquartile range = 45 – 29 = 16
xi x
2
3318
d. s2 174.6316
n 1 20 1
s s 2 174.6316 13.2148
s 13.2148
e. Coefficient of variation = 100% 100% 34.8%
x 38
xi 260
60. a. x 18.57
n 14
Median = 16.5 (Average of 7th and 8th values)
b. s2 = 53.49 s = 7.31
c. Quantex has the best record: 11 Days
27 18.57
d. z 115
.
7.31
PackardBell is 1.15 standard deviations slower than the mean.
12 18.57
e. z 0.90
7.31
IBM is 0.9 standard deviations faster than the mean.
f. Check Toshiba:
37 18.57
z 2.52
7.31
On the basis of z scores, Toshiba is not an outlier, but it is 2.52 standard deviations slower than the
mean.
3 27
Chapter 3
62. a. The sample mean is 770,010.88 and the sample standard deviation is 138,308.23.
b. With z = 2, Chebyshev’s theorem gives:
1 1 1 3
1 2
1 2 1
z 2 4 4
Therefore, at least 75% of household incomes are within 2 standard deviations of the mean. Using
the sample mean and sample standard deviation computed in part (a), the range within 75% of
household incomes must fall is 770,010.88 2(138,308.23) = 770,010.88 276,616.46 ; thus, 75%
of household incomes must fall between 493,394 and 1,046,627 or NT$493,394 to NT$1,046,627.
c. With z = 2, the empirical rule suggests that 95% of household incomes must fall between
NT$493,394 to NT$1,046,627. For the same range, the percentage obtained using the empirical rule
is greater than the percentage obtained using Chebyshev’s theorem.
d. The disposable income for Hsinchu City is 1,078,255; thus, Hsinchu City observation is an outlier.
320
63. a. Public Transportation: x 32
10
320
Automobile: x 32
10
b. Public Transportation: s = 4.64
Automobile: s = 1.83
c. Prefer the automobile. The mean times are the same, but the auto has less variability.
d. Data in ascending order:
Public: 25 28 29 29 32 32 33 34 37 41
Auto: 29 30 31 31 32 32 33 33 34 35
Five number Summaries
Public: 25 29 32 34 41
3 28
Descriptive Statistics: Numerical Methods
Auto: 29 31 32 33 35
Box Plots:
Public:
24 28 32 36 40
Auto:
24 28 32 36 40
The box plots do show lower variability with automobile transportation and support the conclusion
in part c.
64. a. The sample covariance is 2,614,606,406.32. Because the sample covariance is positive, there is a
positive linear relationship between disposable income and rent.
b. The sample correlation coefficient is .928; this indicates a strong linear relationship between
disposable income and rent.
65. a. Let x = media expenditures ($ millions) and y = shipments in barrels (millions)
404.1 119.9
xi 404.1 x 40.41 yi 119.9 y 11.99
10 10
( xi x )( yi y ) 3763.481
sxy 418.1646
n 1 10 1
A positive relationship
( xi x )2 19, 248.469
b. sx 46.2463
n 1 10 1
3 29
Chapter 3
( yi y ) 2 939.349
sy 10.2163
n 1 10 1
sxy 418.1646
rxy 0.885
sx s y (46.2463)(10.2163)
Note: The same value can also be obtained using Excel's CORREL function
66. a. The scatter diagram indicates a positive relationship
Strong positive relationship
67. a. The scatter diagram is shown below:
3.5
2.5
Earnings
1.5
0.5
0
0 5 10 15 20 25 30
Book Value
b. The sample correlation coefficient is .75; this indicates a positive linear relationship between book
value and earnings.
68. a. (800 + 750 + 900)/3 = 817
3 30
Descriptive Statistics: Numerical Methods
70.
fi Mi f i Mi Mi x ( Mi x ) 2 fi ( Mi x ) 2
10 47 470 13.68 187.1424 1871.42
40 52 2080 8.68 75.3424 3013.70
150 57 8550 3.68 13..5424 2031.36
175 62 10850 +1.32 1.7424 304.92
75 67 5025 +6.32 39.9424 2995.68
15 72 1080 +11.32 128.1424 1922.14
10 77 770 +16.32 266.3424 2663.42
475 28,825 14,802.64
28,825
a. x 60.68
475
14,802.64
b. s2 31.23
474
s 31.23 5.59
71.
fi Mi f i Mi Mi x ( Mi x ) 2 fi ( Mi x ) 2
2 29.5 59.0 22 484 968
6 39.5 237.0 12 144 864
4 49.5 198.0 2 4 16
4 59.5 238.0 8 64 256
2 69.5 139.0 18 324 648
2 79.5 159.0
28 784 1568
20 1,030.0 4320
1030
x 51.5
20
4320
s 227.37
19
s = 15.08
3 31