Professional Documents
Culture Documents
THE PRACTICE
OF BUSINESS
STATISTICS
DAVID S. MOORE
GEORGE P. MCCABE
Purdue University
WILLIAM DUCKWORTH
Iowa State University
STANLEY SCLOVE
University of Illinois at Chicago
Answers 39
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 1
It is common to make boxplots of large data sets using the 5% and 95% points in place
of the minimum and maximum. The highest income among the 31,970 people with only
a high school education, for example, is $425,510. It is more informative to see that 95%
of this group earned less than $56,294. The 5% and 95% points contain between them the
middle 90% of the observations.
(a) Use this output to make boxplots that compare the income distributions for the five
education groups.
(b) Write a brief summary of the relationship between education and income. For example,
do people who start college but don’t get a degree do much better than people with only a
high school education?
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 3
Table 1.10
Monthly percent returns on Phillip Morris stock
from August 1990 to August 2001
3.0 -5.7 1.2 4.1 3.2 7.3 7.5 18.7 3.7 -1.8
2.4 -6.5 6.7 9.4 -2.0 -2.8 -3.4 19.2 -4.8 0.5
-0.6 2.8 -0.5 -4.5 8.7 2.7 4.1 -10.3 4.8 -2.3
-3.1 -10.2 -3.7 -26.6 7.2 -2.4 -2.8 3.4 -4.6 17.2
4.2 0.5 8.3 -7.1 -8.4 7.7 -9.6 6.0 6.8 10.9
1.6 0.2 -2.4 -2.4 3.9 1.7 9.0 3.6 7.6 3.2
-3.7 4.2 13.2 0.9 4.2 4.0 2.8 6.7 -10.4 2.7
10.3 5.7 0.6 -14.2 1.3 2.9 11.8 10.6 5.2 13.8
-14.7 3.5 11.7 1.5 2.0 -3.2 -3.9 -4.7 9.8 4.9
-8.3 4.8 -3.2 -10.9 0.7 6.4 11.3 -5.1 12.3 10.5
9.4 -3.6 -12.4 -16.5 -8.9 -0.4 10.0 5.4 -7.3 0.5
-7.4 -22.9 -0.5 -10.6 -9.2 -3.3 5.2 5.4 19.4 3.5
-4.9 17.8 0.7 24.4 4.3 16.6 0.0 9.5 -0.4 5.6
2.6 -2.7 -8.1 4.2
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 5
Figure 2.7
80
o
o o o o
o o o oo o
o Costa Rica o o ooo o o
o oo
o o o o o USA
o o
oo o
o o o o
o o oo o
70
oo oo o o o
o oo o o
oo o
o o o o
o o
Life expectancy, years
oo o
o
o
o
o
60
o
o o
oo o
o o Namibia
o
ooo
ooo
o oo o Gabon
oo
50
o
o
o o
o o o Botswana
o o
oo
oo
ooo
oo
40
Make a scatterplot suitable for predicting fund returns from EAFE returns. Is there a clear
straight-line pattern? How strong is this pattern? (Give a numerical measure.) Are there
any extreme outliers from the straight-line pattern?
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 15
(a) What is the equation of the least-squares line? Use this line to predict the percent
return on stocks in a year when Treasury bills return 5%.
(b) Explain what the slope of the regression line tells us. Does the slope confirm that high
interest rates are in general bad for stocks?
(c) If you knew the return on Treasury bills for next year, do you think you could predict
the return on stocks quite accurately? Use both the scatterplot in Figure 2.25 and the
regression output to justify your answer.
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 23
Figure 2.25
60
o
40 o
oo
o o
o o
o o o
o
Stock return (percent)
o oo o
o o
o o o
20
oo o o
o o o
o o
o
o o o
o o
o o o o
0
o o
o
o o o
oo o
o
-20
o
-40
0 5 10 15 20
Treasury bill return (percent)
24 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
Selling Price Square Footage ($) Age (as of 2000) Selling Price Square Footage ($) Age (as of 2000)
268380 1897 1 169900 1686 35
131000 1157 15 180000 2054 34
112000 1024 35 127000 1386 50
112000 935 35 242500 2603 10
122000 1236 39 152900 1582 3
127900 1248 32 171600 1790 1
157600 1620 33 195000 1908 6
135000 1124 33 83100 1378 72
145900 1248 35 125000 1668 55
126000 1139 39 60500 1248 100
142000 1329 40 85000 1229 59
107500 1040 45 117000 1308 60
110000 951 42 57000 892 90
187000 1628 1 110000 1981 72
94000 816 43 127250 1098 70
99500 1060 24 119000 1858 80
78000 800 68 172500 2010 60
55790 492 79 123000 1680 86
70000 792 80 161715 1670 1
53600 980 62 179797 1938 1
157000 1629 3 117250 1120 36
166730 1889 0 116500 914 4
340000 2759 6 117000 1008 23
195000 1811 3 177500 1920 32
215850 2400 27 132000 1146 37
26 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
Color Silver White Black Dark Green Dark Blue Medium Red
Probability 0.176 0.172 0.113 0.089 0.088 0.067
What is the probability that the car you choose has any color other than the six listed?
What is the probability that a randomly chosen car is either silver or white?
32 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
(a) Verify that this is a legitimate discrete distribution. Display the distribution in a
probability histogram.
(b) Say in words what the event {X ≥ 1} is. Find P (X ≥ 1).
(c) Your company builds houses with two-car garages. What percent of households have
more cars than the garage can hold?
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 33
Rooms 1 2 3 4 5 6 7 8 9 10
Owned .003 .002 .023 .104 .210 .224 .197 .149 .053 .035
Rented .008 .027 .287 .363 .164 .093 .039 .013 .003 .003
(a) Make probability histograms of these two distributions, using the same scales. What
are the most important differences between the distributions for owner-occupied and rented
housing units?
(b) Find the mean number of rooms for both types of housing unit. How do the means
reflect the differences you found in (a)?
34 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
Portfolio analysis
Here are the means, standard deviations, and correlations for the monthly returns from
three Fidelity mutual funds for the 36 months ending in December 2000. Because there are
now three random variables, there are three correlations. We use subscripts to show which
pair of random variables a correlation refers to.
W = monthly return on Magellan Fund µW = 1.14% σW = 4.64%
X = monthly return on Real Estate Fund µX = 0.16% σX = 3.61%
Y = monthly return on Japan Fund µY = 1.59% σY = 6.75%
Correlations
ρW X = 0.19 ρW Y = 0.54 ρXY = −0.17
Answers to Problems
selected from
Chapters 1−4 for
THE PRACTICE OF BUSINESS STATISTICS
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 41
10
5
0
Ag. Mines Const. Manuf. Trans. Whole- Retail Finance Svc. Govt. Other
& sale
Utils.
42 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
10
5
0
Const. Trans. Ag. Svc. Manuf. Govt. Retail Whole- Mines Finance Other
& sale
Utils.
(d) Yes it would because each death can be assigned to exactly one category.
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 43
150000
Income
50000
0
(b) Median income in March of 2000 appeared to increase with education level, although there
was more income variability (as evidenced by the IQR) in each progressively higher education
group. On average those with a higher level of education earn more, but there are plenty of
exceptions to this pattern. For example, the top 25% of earners with only a high school diploma
earn more than the bottom 25% with a Master's degree.
44 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
20
10
0
Percent return
(b) The outliers are -26.6 and -22.9. After the outliers have been omitted, the distribution of
monthly returns is unimodal and approximately symmetric with only some left skew. The center
of the returns appears to be just above zero. The returns range from about -20 to about 25, with
well over half of the returns between -10 and 15.
(c) The mean is 1.63, the standard deviation is 8.29. An investment of $100 would be worth
$101.63.
(d) $73.4. Without the outliers, the mean is 2.04 and the standard deviation is 7.65. The mean
increases by about .41 percentage points, while the standard deviation drops by 0.63 percentage
points. Since the median and quartiles are robust or resistant to outliers, they would not change
very much if the outliers were omitted.
CA
15000
Business failures
10000
TX
5000
NY
FL
0
Business starts
(b) Because, on average, as the explanatory variable increases, so does the response or dependent
variable. Further, the straight line that might be used to describe these data has a positive slope.
(c) See labeled point on plot above.
(d) California.
(e) The states are California, Florida, New York, and Texas. They are not from one region, but
are the four most populous states.
46 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
CA
100
Target stores
TX
50
0
Wal-Mart stores
(b) Two points stand out: California and Texas. California has a very high number of Target
stores, and a relatively high number of Wal-Marts. Texas has an extremely high number of Wal-
Mart stores and more Targets than every state but California.
(c) The relationship between Wal-Mart stores and Target stores is positive and approximately
linear. The strength of the relationship is moderate to weak since the points are not tightly
clustered about a line.
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 47
30
20
10
0
0 20 40 60
= credit
800
= cash
600
Gross sales ($)
400
200
0 10 20 30 40 50
Items sold
(b) Gross sales = 240.88 + 10.34(Item count credit card); Gross sales = 64.58 + 12.46(Item count
cash); One additional cash purchase is expected to increase the daily gross sales by an estimated
$12.46, while an additional credit card purchase is expected to increase the gross sales by an
estimated $10.34.
(b) There appears to be a moderate to strong positive linear relationship between the square
footage of an Ames-area home and its selling price. See plot below.
350000
250000 Selling price vs. square feet of Ames, IA homes
Price ($)
150000
50000
Square feet
(c) The estimated intercept is 4786.46 and the estimated slope is 92.82. On average, an additional
square foot will add an estimated $92.82 to the selling price of a home.
(d) $153,300.10.
(e) 69.64%.
For the matched pairs design, ten subject are initially assigned computers with the trend
highlighting software while the other ten are given standard computers. After a fixed trading
period, students switch computer types and trade for the same amount of time. One possible
randomization is to initially assign students C, D, F, H, I, J, L, M, N, and S to the computers with
the trend software and students A, B, E, G, K, O, P, Q, R, and T to the computers without it.
0.4
0.3 Probability histogram for number of cars owned
Probability
0.2
0.1
0.0
0 1 2 3 4 5
Cars
(b) A randomly chosen household has one or more cars. The probability of this event is 0.91.
(c) 0.20.
Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics 53
0.2
0.1
0.0
1 2 3 4 5 6 7 8 9 10
Rooms
54 Moore/McCabe/Duckworth/Sclove: The Practice of Business Statistics
0.4
0.3
Probability
0.2
0.1
0.0
1 2 3 4 5 6 7 8 9 10
Rooms
(b) The mean for owner-occupied units is 6.284, the mean for renter-occupied units is 4.187. The
difference in means is evident in the plots.