You are on page 1of 31

Chapter 3 Descriptive Statistics: Numerical Methods

Learning Objectives 1. 2. 3. 4. Understand the purpose of measures of location. Be able to compute the mean, median, mode, quartiles, and various percentiles. Understand the purpose of measures of variability. Be able to compute the range, interquartile range, variance, standard deviation, and coefficient of variation. 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. Understand how z scores are computed and how they are used as a measure of relative location of a data value. Know how Chebyshevs 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. Learn how to construct a 5-number summary and a box plot. Be able to compute and interpret covariance and correlation as measures of association between two variables. Be able to compute a weighted mean.

5.

6.

7.

8. 9.

10.

3-1

Chapter 3

Solutions: x= xi 75 = = 15 n 5

1.

10, 12, 16, 17, 20 Median = 16 (middle value) x= xi 96 = = 16 n 6

2.

10, 12, 16, 17, 20, 21 Median = 16 + 17 = 16.5 2

3.

15, 20, 25, 25, 27, 28, 30, 34 i= 20 (8) = 1.6 100 25 (8) = 2 100 65 (8) = 5.2 100 75 (8) = 6 100

2nd position = 20 20 + 25 = 22.5 2

i=

i=

6th position = 28 28 + 30 = 29 2

i=

4.

Mean =

xi 657 = = 59.727 n 11 6th item

Median = 57

3-2

Descriptive Statistics: Numerical Methods

Mode = 53

It appears 3 times

5.

a.

x=

x
n

1194.50 = 39.82 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. Median =
40.2 + 40.8 = 40.5 2

c. d.

Mode = 42.0 This value appears 3 times


25 30 = 7.5 100 Rounding up, we see that Q1 is at the 8th position.

First Quartile i =

Q1 = 38.2 Third Quartile i =

e.

75 30 = 22.5 100

Rounding up, we see that Q3 is at the 23rd position. Q3 = 42.1 x= xi 871.74 = = 36.32 n 24

6.

a.

Median is average of 10th and 11th values after arranging in ascending order. Median = 39.00 + 39.95 = 39.48 2

Data are multimodal x= xi 491.14 = = 20.46 n 24 19.75 + 19.95 = 19.85 2

b.

Median =

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 broker-assisted 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 broker-assisted or online trades, we would probably find that larger transactions cost more. x= xi 1380 = = 46 n 30

7.

a.

b. c.

Yes, the mean here is 46 minutes. The newspaper reported an average of 45 minutes. Median = 45 + 52.9 = 48.95 2

d.

Q1 = 7 (value of 8th item in ranked order) Q3 = 70.4 (value of 23rd item in ranked list)

e.

40 Find position i = 30 = 12; 40th percentile is average of values in 12th and 13th positions. 100 40th percentile = xi = 695 x= xi 695 = = 34.75 n 20 28.8 + 29.1 = 28.95 2

8.

a.

The modal age is 29; it appears 3 times. b. Median is average of 10th and 11th items. Median = 33 + 36 = 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 = 25.5 2

Q1 =

For Q3, 75 i= 20 = 15 100 Since i is integer, Q3 = 42 + 45 = 43.5 2

d.

32 i= 20 = 6.4 100 Since i is not an integer, we round up to the 7th position. 32nd percentile = 27

9.

a.

x=

xi 270,377 = = 10,815.08 Median (Position 13) = 8296 n 25

b. c.

Median would be better because of large data values. 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 x= xi 435 = = 48.33 n 9

Data in ascending order:

3-5

Chapter 3

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 i= 9 = 2.25 100 Q1 (3rd position) = 45 75 i= 9 = 6.75 100 Q3 (7th position) = 55 11. Using the mean we get xcity =15.58, xcountry = 18.92 For the samples we see that the mean mileage is better in the country than in the city.

b.

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 12, 780 = = $639 n 20 xi 1976 = = 98.8 pictures n 20 xi 2204 = = 110.2 minutes n 20

12. a.

x=

b.

x=

c.

x=

d.

This is not an easy choice because it is a multicriteria problem. If price was the only criterion, the lowest price camera (Fujifilm DX-10) 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. Range 20 - 10 = 10 10, 12, 16, 17, 20 i= 25 (5) = 1.25 100

13.

Q1 (2nd position) = 12 i= 75 (5) = 3.75 100

Q3 (4th position) = 17 IQR = Q3 - Q1 = 17 - 12 = 5

14.

x=

xi 75 = = 15 n 5 ( xi x ) 2 64 = = 16 n 1 4

s2 =

s = 16 = 4 15. 15, 20, 25, 25, 27, 28, 30, 34 Range = 34 - 15 = 19

3-7

Chapter 3 25 (8) = 2 100 75 (8) = 6 100 20 + 25 = 22.5 2 28 + 30 = 29 2

i=

Q1 =

i=

Q3 =

IQR = Q3 - Q1 = 29 - 22.5 = 6.5 x= xi 204 = = 255 . n 8 ( xi x ) 2 242 = = 34.57 n 1 7

s2 =

s = 34.57 = 588 . 16. a. b. Range = 190 - 168 = 22 ( xi x ) 2 = 376 s 2 = 376 = 75.2 5 c. s = 75.2 = 8.67 8.67 Coefficient of Variation = 100% = 4.87% 178 Range = 92-67 = 25 IQR = Q3 - Q1 = 80 - 77 = 3 x = 78.4667

d.

17.

( x x )
i

= 411.7333
2

( x x ) =
i

n 1

411.7333 = 29.4095 14

18. a.

s = 29.4095 = 5.4231 x 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 86.24 26.82 23.30% Asia 42.97 11.40 31.13%

Range Standard Deviation Coefficient of Variation c. 19. a.

Greater mean and standard deviation for Mainland. Greater coefficient of variation for Asia. Range = 60 - 28 = 32 IQR = Q3 - Q1 = 55 - 45 = 10

b.

x=

435 = 48.33 9

( xi x ) 2 = 742 s2 = ( xi x )2 742 = = 92.75 n 1 8

s = 92.75 = 9.63 c. 20. The average air quality is about the same. But, the variability is greater in Anaheim. Dawson Supply: Range = 11 - 9 = 2 s= 4.1 = 0.67 9

J.C. Clark: Range = 15 - 7 = 8 s= 60.1 = 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. Winter Summer 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. 22. a. More variability in the summer months. 100 Shares at $50 (Broker-assisted) Min Value = 9.95 Max Value = 55.00 Variance 8.2333 44.4889 Standard Deviation 2.8694 6.6700

Range = 55.00 - 9.95 = 45.05 Q1 = 24.99 + 25.00 = 24.995 2 Q3 = 48.00 + 49.95 = 48.975 2

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 Q1 = 12.95 + 14.00 = 13.475 2 Q3 = 24.95 + 24.95 = 24.95 2

Interquartile range = 24.950 - 13.475 = 11.475 b. 100 Shares at $50 (Broker-assisted)

3 - 10

Descriptive Statistics: Numerical Methods ( xi x )2 = 190.67 n 1

s2 =

s = s 2 = 13.81 500 Shares at $50 (Online) s2 = ( xi x )2 = 140.63 n 1

s = s 2 = 11.86 c. 100 Shares at $50 (Broker-assisted) s 13.81 (100%) = (100%) = 38.02% x 36.32

Coefficient of Variation =

500 Shares at $50 (Online) s 11.86 (100%) = (100%) = 57.97% x 20.46

Coefficient of Variation =

d.

Using the standard deviation as a measure, the variability seems to be greater for the broker-assisted trades. But, using the coefficient of variation as a measure, we see that the relative variability is greater for the online trades. s2 = 0.0021 Production should not be shut down since the variance is less than .005. 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. x= xi 75 = = 15 n 5

23. 24.

25.

3 - 11

Chapter 3

s2 =

( xi x ) 2 = n 1 z=

64 =4 4

10

10 15 = 1.25 4 20 15 = +1.25 4

20

z=

12 17

12 15 = 0.75 4 17 15 z= = +.50 4 z= z= 16 15 = +.25 4

16

26.

z=

520 500 = +.20 100 650 500 = +1.50 100 500 500 = 0.00 100 450 500 = 0.50 100 280 500 = 2.20 100 40 30 =2 5 45 30 =3 5 38 30 = 1.6 5 42 30 = 2.4 5 48 30 = 3.6 5 1 1 = 0.75 At least 75% 22 1 = 0.89 At least 89% 32 1 = 0.61 At least 61% 1.62 1 = 0.83 At least 83% 2.42 1 = 0.92 At least 92% 3.6 2

z=

z=

z=

z=

27. a.

z=

b.

z=

c.

z=

d.

z=

e.

z=

3 - 12

Descriptive Statistics: Numerical Methods

28. a. b. c. 29. a.

Approximately 95% Almost all Approximately 68% This is from 2 standard deviations below the mean to 2 standard deviations above the mean. With z = 2, Chebyshevs theorem gives: 1 1 1 1 3 = 1 2 = 1 = 2 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. With z = 2.5, Chebyshevs theorem gives: 1 1 1 1 = 1 = 1 = .84 2 2 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 Chebyshevs theorem. 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 bell-shaped distribution is symmetric, approximately, 34% of the kids watch television between 2 and 3 hours per day. 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 bell-shaped 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. 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. Approximately 68% of scores are within 1 standard deviation from the mean. Approximately 95% of scores are within 2 standard deviations from the mean. Approximately (100% - 95%) / 2 = 2.5% of scores are over 130. Yes, almost all IQ scores are less than 145.

30. a.

b.

c.

31. a. b. c. d.

3 - 13

Chapter 3

32. a.

z=

71.00 90.06 = 0.95 20 168 90.06 = 3.90 20

b.

z=

c.

The z-score in part a indicates that the value is 0.95 standard deviations below the mean. The zscore 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.

33. a. b.

x is approximately 43.48 or NT$43,480, and s is 3.45 or NT$3,450


This is from 2 standard deviations below the mean to 2 standard deviations above the mean. With z = 2, Chebyshevs theorem gives: 1 1 1 1 3 = 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:

16 14 12 10
Frequency

8 6 4 2 0 36-38 38-40 40-42 42-44


S alary

44-46

46-48

48-50

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. d. With z = 2, the empirical rule suggests that 95% of industry and service workers have an annual salary between NT$36,580 and NT$50,380. The percentage is much higher than obtained using Chebyshevs theorem, but requires the assumption that the distribution of monthly salary is bell shaped. There are no outliers because all the observations are within 3 standard deviations of the mean.

e. 34. a. b.

x is 100 and s is 13.88 or approximately 14


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 bell-shaped distribution is symmetric, approximately 16% of the winning teams will score more than 114 points. 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 z-score for the winning margin that is farthest from the sample mean of 11.1, a winning margin of 32 points. z= x x 32 11.1 = = 1.94 s 10.77

c.

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 z-score that is less than 3 or greater than 3 and hence we conclude that there are no outliers x= xi 79.86 = = 3.99 n 20 4.17 + 4.20 = 4.185 (average of 10th and 11th values) 2

35. a.

Median =

b.

Q1 = 4.00 (average of 5th and 6th values) Q3 = 4.50 (average of 15th and 16th values) s= ( xi x ) 2 12.5080 = = 0.8114 n 1 19

c.

3 - 15

Chapter 3

d. e.

The distribution is significantly skewed to the left. Allison One: z = 4.12 3.99 016 . 0.8114 2.32 3.99 2.06 0.8114

Omni Audio SA 12.3: z =

f.

The lowest rating is for the Bose 501 Series. Its z-score is: z= 2.14 3.99 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 i= 25 (8) = 2 100 25 + 27 = 26 2 Q3 = 28 + 30 = 29 2 Q1 = 20 + 25 = 22.5 2

Median =

i=

75 (8) = 8 100

Largest = 34

37.

15 38. 5, 6, 8, 10, 10, 12, 15, 16, 18 Smallest = 5 i=

20

25

30

35

25 (9) = 2.25 Q1 = 8 (3rd position) 100

3 - 16

Descriptive Statistics: Numerical Methods

Median = 10 i= 75 (9) = 6.75 Q3 = 15 (7th position) 100

Largest = 18

5 39. IQR = 50 - 42 = 8 Lower Limit: Upper Limit: 65 is an outlier 40. a. b.

10

15

20

Q1 - 1.5 IQR = 42 - 12 = 30 Q3 + 1.5 IQR = 50 + 12 = 62

Five number summary: 5 9.6 14.5 19.2 52.7 IQR = Q3 - Q1 = 19.2 - 9.6 = 9.6 Lower Limit: Upper Limit: Q1 - 1.5 (IQR) = 9.6 - 1.5(9.6) = -4.8 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 41. a.

20

35

50

65

Median (11th position) 4019

3 - 17

Chapter 3

i=

25 (21) = 5.25 100

Q1 (6th position) = 1872 i= 75 (21) = 15.75 100

Q3 (16th position) = 8305 608, 1872, 4019, 8305, 14138 b. Limits: IQR = Q3 - Q1 = 8305 - 1872 = 6433 Lower Limit: Upper Limit: c. d. Q1 - 1.5 (IQR) = -7777 Q3 + 1.5 (IQR) = 17955

There are no outliers, all data are within the limits. 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.

0 42. a. b. c.

3,000

6,000

9,000

12,000

15,000

Mean = 105.7933 Median = 52.7 Q1 = 15.7 Q3 = 78.3 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. 43. a. Mean percentage = 26.73. The current percentage is much greater. 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

50 Small Size

60

70

80

90

100

110

120

130

50 c. 44. a. b. c.

60

70

80

90

100

110

120

130

140

150

The midsize cars appear to be safer than the small cars. x = 37.48 Median = 23.67 Q1 = 7.91 Q3 = 51.92 IQR = 51.92 - 7.91 = 44.01 Lower Limit: Upper Limit: Q1 - 1.5(IQR) = 7.91 - 1.5(44.01) = -58.11 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 x 15 20

45. a. b.

Negative relationship x= 40 =8 5 yi = 230 y= 230 = 46 5 ( yi y ) 2 = 520

c/d. xi = 40

( xi x )( yi y ) = 240 sxy =

( xi x ) 2 = 118

( xi x )( yi y ) 240 = = 60 n 1 5 1 ( xi x ) 2 = n 1 ( yi y ) 2 = n 1 sxy sx s y = 118 = 5.4314 5 1 520 = 11.4018 5 1

sx =

sy =

rxy =

60 = 0.969 (5.4314)(11.4018)

There is a strong negative linear relationship.

3 - 20

Descriptive Statistics: Numerical Methods

46. a.
18 16 14 12 10 8 6 4 2 0 0 5 10 15 x 20 25 30

b.

Positive relationship x= 80 = 16 5 yi = 50 y= 50 = 10 5 ( yi y ) 2 = 86

c/d. xi = 80

( xi x )( yi y ) = 106 sxy =

( xi x ) 2 = 272

( xi x )( yi y ) 106 = = 26.5 n 1 5 1 ( xi x ) 2 = n 1 272 = 8.2462 5 1

sx =

( yi y ) 2 86 = = 4.6368 n 1 5 1 sxy 26.5 rxy = = = 0.693 sx s y (8.2462)(4.6368) sy = 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 x = GPA 3.4 3.6 3.8

b.

Positive relationship x= 19.8 = 3.3 6 yi = 3540 y= 3540 = 590 6

c/d. xi = 19.8

( xi x )( yi y ) = 143 sxy =

( xi x ) 2 = 0.74

( yi y ) 2 = 36,400

( xi x )( yi y ) 143 = = 28.6 n 1 6 1 ( xi x ) 2 = n 1 ( yi y ) 2 = n 1 sxy sx s y = 0.74 = 0.3847 6 1 36, 400 = 85.3229 6 1

sx =

sy =

rxy =

28.6 = 0.8713 (0.3847)(85.3229)

A positive linear relationship 48. Let x = driving speed and y = mileage

3 - 22

Descriptive Statistics: Numerical Methods 420 = 42 10 270 = 27 10 ( y i y ) 2 = 164

xi = 420

x=

yi = 270

y=

( xi x )( yi y ) = 475 sxy =

( xi x ) 2 = 1660

( xi x )( yi y ) 475 = = 52.7778 n 1 10 1 ( xi x ) 2 1660 = = 13.5810 n 1 10 1 ( yi y ) 2 164 = = 4.2687 n 1 10 1 s xy sx s y = 52.7778 = .91 (13.5810)(4.2687)

sx =

sy =

rxy =

A strong negative linear relationship 49. a. b. 50. a. b. 51. The sample correlation coefficient is .78. There is a positive linear relationship between the performance score and the overall rating. The sample correlation coefficient is .92. There is a strong positive linear relationship between the two variables. The sample correlation coefficient is .88. This indicates a strong positive linear relationship between the daily high and low temperatures. x= wi xi 6(3.2) + 3(2) + 2(2.5) + 8(5) 70.2 = = = 3.69 wi 6+ 3+ 2 +8 19

52. a.

b. 53.

3.2 + 2 + 2.5 + 5 12.7 = = 3175 . 4 4

fi 4 7 9 5 25 x= f i M i 325 = = 13 n 25

Mi 5 10 15 20

fi M i 20 70 135 100 325

3 - 23

Chapter 3 Mi x -8 -3 +2 +7 (M i x )2 64 9 4 49 f i ( M i x )2 256 63 36 245 600

fi 4 7 9 5

Mi 5 10 15 20

s2 =

f i ( M i x ) 2 600 = = 25 n 1 24

s = 25 = 5 54. a. Grade xi 4 (A) 3 (B) 2 (C) 1 (D) 0 (F) Weight Wi 9 15 33 3 0 60 Credit Hours

x=

wi xi 9(4) + 15(3) + 33(2) + 3(1) 150 = = = 2.50 wi 9 + 15 + 33 + 3 60

b.

Yes; satisfies the 2.5 grade point average requirement x= wi xi 3000(15) + 5500(14) + 4200(12) + 3000(25) + 3000(20) + 3800(12) + 2500(35) = wi 3000 + 5500 + 4200 + 3000 + 3000 + 3800 + 2500 440,500 = 17.62% 25, 000 wi xi 3000(1.21) + 5500(1.48) + 4200(1.72) + 3000(0) + 3000(.96) + 3800(2.48) + 2500(0) = wi 3000 + 5500 + 4200 + 3000 + 3000 + 3800 + 2500 31, 298 = 1.25% 25, 000

55. a.

b.

x=

56. Mi 2 7 12 17 22 27 fi 74 192 280 105 23 6 680 fi Mi 148 1,344 3,360 1,785 506 162 7,305 Mi x -8.742647 -3.742647 1.257353 6.257353 11.257353 16.257353 (M i x )2 76.433877 14.007407 1.580937 39.154467 126.728000 264.301530 fi ( M i x )2 5,656.1069 2,689.4221 442.6622 4,111.2190 2,914.7439 1,585.8092 17,399.9630

3 - 24

Descriptive Statistics: Numerical Methods

x=

7305 = 10.74 680 17,399.9630 = 25.63 679

s2 =

s = 5.06 Estimate of total gallons sold: (10.74)(120) = 1288.8 57. a. Class 0 1 2 3 4 Totals x= i fM i 1745 = = 3.49 n 500 fi 15 10 40 85 350 500 Mi 0 1 2 3 4 fi Mi 0 10 80 255 1400 1745

b. Mi x -3.49 -2.49 -1.49 -0.49 +0.51 ( Mi x )2 12.18 6.20 2.22 0.24 0.26 Total fi ( Mi x )2 182.70 62.00 88.80 20.41 91.04 444.95

s2 =

( M i x ) 2 f i 444.95 = = 0.8917 n 1 499

s = 0.8917 = 0.9443

58. a.

x=

x
n

37640 = 1505.6 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. Q1 = 982 (7th value)

c.

3 - 25

Chapter 3

Q3 = 2100 (19th value)

d.

Min = 0

Max = 3280

Range = 3280 - 0 = 3280 IQR = Q3 - Q1 = 2100 - 982 = 1118 e. f. s2 = 847722.3 s = 920.72

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. The z - score for the largest value is:

g.

z=

3280 1505.6 = 1.93 920.72

It is the only outlier and should be checked for accuracy. 59. a. xi = 760 x= xi 760 = = 38 n 20

Median is average of 10th and 11th items. Median = 36 + 36 = 36 2

The modal cash retainer is 40; it appears 4 times. b. For Q1, 25 i= 20 = 5 100 Since i is integer, Q1 = 28 + 30 = 29 2

For Q3,

3 - 26

Descriptive Statistics: Numerical Methods 75 i= 20 = 15 100 Since i is integer, Q3 = c. 40 + 50 = 45 2

Range = 64 15 = 49 Interquartile range = 45 29 = 16 s2 = ( xi x ) n 1


2

d.

3318 = 174.6316 20 1

s = s 2 = 174.6316 = 13.2148 s 13.2148 Coefficient of variation = 100% = 100% = 34.8% x 38 x= xi 260 = = 18.57 n 14

e.

60. a.

Median = 16.5 (Average of 7th and 8th values) b. c. d. s2 = 53.49 s = 7.31

Quantex has the best record: 11 Days z= 27 18.57 = 115 . 7.31

Packard-Bell is 1.15 standard deviations slower than the mean. e. z= 12 18.57 = 0.90 7.31

IBM is 0.9 standard deviations faster than the mean. f. Check Toshiba: z= 37 18.57 = 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

61.

Sample mean = 7195.5 Median = 7019 (average of positions 5 and 6) Sample variance = 7,165,941 Sample standard deviation = 2676.93

62. a. b.

The sample mean is 770,010.88 and the sample standard deviation is 138,308.23. With z = 2, Chebyshevs theorem gives: 1 1 1 1 3 = 1 2 = 1 = 2 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 Chebyshevs theorem. The disposable income for Hsinchu City is 1,078,255; thus, Hsinchu City observation is an outlier. Public Transportation: x = Automobile: x = 320 = 32 10 320 = 32 10

d. 63. a.

b.

Public Transportation: s = 4.64 Automobile: s = 1.83

c. d.

Prefer the automobile. The mean times are the same, but the auto has less variability. Data in ascending order: Public: Auto: 25 28 29 29 32 32 33 34 37 41 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: Box Plots:

29 31 32 33 35

Public:

24 Auto:

28

32

36

40

24

28

32

36

40

64. a.

The box plots do show lower variability with automobile transportation and support the conclusion in part c. 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. The sample correlation coefficient is .928; this indicates a strong linear relationship between disposable income and rent.

b.

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 = ( xi x ) 2 = 19, 248.469 ( y i y ) 2 = 939.349

( xi x )( yi y ) 3763.481 = = 418.1646 n 1 10 1

A positive relationship sx = ( xi x ) 2 19, 248.469 = = 46.2463 n 1 10 1

b.

3 - 29

Chapter 3

sy =

( yi y ) 2 939.349 = = 10.2163 n 1 10 1 sxy sx s y = 418.1646 = 0.885 (46.2463)(10.2163)

rxy =

Note: The same value can also be obtained using Excel's CORREL function 66. a. b. The scatter diagram indicates a positive relationship xi = 798 yi = 11, 688 xi yi = 1, 058, 019

xi2 = 71,306 rxy =

yi2 = 16, 058, 736 = 1, 058, 019 (798)(11, 688) / 9 71,306 (798) 2 / 9 16, 058, 736 (11, 688) 2 / 9 = .9856

xi yi ( xi yi ) / n xi2 ( xi ) / n yi2 ( yi ) / n
2 2

Strong positive relationship

67. a.

The scatter diagram is shown below:


3.5 3 2.5 Earnings 2 1.5 1 0.5 0 0 5 10 15 Book Value 20 25 30

b.

The sample correlation coefficient is .75; this indicates a positive linear relationship between book value and earnings. (800 + 750 + 900)/3 = 817 Month Weight January 1 February 2 March 3

68. a. b.

3 - 30

Descriptive Statistics: Numerical Methods wi xi 1(800) + 2(750) + 3(900) 5000 = = = 833 wi 1+ 2 + 3 6 wi xi 20(20) + 30(12) + 10(7) + 15(5) + 10(6) 965 = = = 11.4 days wi 20 + 30 + 10 + 15 + 10 85

x=

69. 70.

x=

fi 10 40 150 175 75 15 10 475 x=

Mi 47 52 57 62 67 72 77

f i Mi 470 2080 8550 10850 5025 1080 770 28,825

Mi x -13.68 -8.68 -3.68 +1.32 +6.32 +11.32 +16.32

( Mi x )2 187.1424 75.3424 13..5424 1.7424 39.9424 128.1424 266.3424

fi ( Mi x )2 1871.42 3013.70 2031.36 304.92 2995.68 1922.14 2663.42 14,802.64

a.

28,825 = 60.68 475

b.

s2 =

14,802.64 = 31.23 474

s = 31.23 = 5.59 71. fi 2 6 4 4 2 2 20 x= 1030 = 51.5 20 4320 = 227.37 19 Mi 29.5 39.5 49.5 59.5 69.5 79.5 f i Mi 59.0 237.0 198.0 238.0 139.0 159.0 1,030.0 Mi x -22 -12 -2 8 18 28 ( Mi x )2 484 144 4 64 324 784 fi ( Mi x )2 968 864 16 256 648 1568 4320

s=

s = 15.08

3 - 31

You might also like