Exercise 1

A sample of 25 families was asked how many pets they owned. Their responses are
summarized in the following table.

Number of Pets 0 1 2 3 4 5
Number of Families 3 10 5 4 2 1

With respect to this sample we have:

1. The mean number of pets owned per family is 2.5.

2. The median number of pets owned per family is 2.5.

Exercise 2

The following data represent the number of children in a sample of 10 families from
Rotterdam: 4, 2, 1, 1, 5, 3, 0, 2, 0, and 2. With respect to this sample we have:

3. The IQR for the number of children is 2.

4. The median number of children equals the mean number of children

5. The standard deviation for the number of children is, rounded to two decimals accuracy,
2.67.

Exercise 3

It is know that the lifetime for a certain type of lamp, say type A, is normally distributed with
a mean of 5000 hours and a standard deviation of 300 hours.

6. The proportion of these lamps that lasts between 4,700 and 5525 hours, rounded to whole
numbers, is 80%.

The manufacturer wishes to advertise a lifetime for these lamps in such a way that only 1.5%
of the lamps will burn out before the advertised lifetime.

7. The advertised lifetime should be, rounded to whole hours, 4349 hours.

To check the quality of the lamps somebody measures the lifetime for 9 lamps.

8. The probability that the average lifetime for these 9 lamps is less than 4850 hours is,
rounded to four decimals accuracy, 0.3085.

For a different type of lamp, say a type B lamp, it is known that the lifetime is also normally
distributed. Furthermore, 22.66% of these lamps last more than 4000 hours whereas 30.85%
last less than 3600 hours.

9. The mean lifetime for the type B lamps is, rounded to whole hours, 3760 hours.

3
Exercise 4
The general manager of a chain of furniture stores believes that experience is the most
important factor in determining the level of success of a salesperson. To examine this belief
she
selected salespeople working for this store. These data are listed below.

Salesperson Years of Experience Sales

(x) (y)
1 0 7
2 2 9
3 10 20
4 3 15
5 8 18
6 5 14
7 12 20
8 7 17
9 20 30
10 15 25

Some descriptive statistics for these data are: sx = 6.2147, sy = 6.8840 , x = 8.2, y = 17.5.
Furthermore, she finds as least-squares regression equation: y 8.63 + 1.0817x.

10. The correlation coefficient between Years of Experience (x) and Sales (y) is, rounded to
two decimals accuracy, 0.98.

11. A salesperson with 10 years of experience indicates that he may be willing to change jobs.
The estimated monthly sales for this person are $20,000.

12. The residual corresponding to salesperson 1 (with no years of experience) is $ 8,630.

Exercise 5
considers both the
expected payoff of the portfolio as well as the risk associated with her portfolio, where she
uses the standard deviation of the payoff as measure for the risk.
she can choose from two stocks: A and B. For each of these stocks, the purchase pric
100.00. The expected pay 10.00. The standard
.00 and for s offs
for stock A and B are independent.

13. When she buys two stocks B, the standard deviation of the payoff of the portfolio will be

14. In order to minimize the risk of the portfolio, she should purchase 1 stock A and 1 stock B.

4
Exercise 6

The joint probability distribution of the random variables X and Y is shown in the table below,
where X is the number of tennis racquets and Y is the number of golf clubs sold daily in a
small sports store.

X
Y 1 2 3
1 0.30 0.18 0.12
2 0.15 0.09 0.06
3 0.05 0.03 0.02

15. X and Y are independent.

16. The standard deviation of the number of sold tennis racquets is, rounded to two decimals
accuracy, 0.78.

17. As the sales for tennis racquets and golf clubs are positively correlated, the covariance
between the number of sold tennis racquets and golf clubs is positive. That is, COV(X,Y)
> 0.

18. The expected value of X + Y cannot be calculated without additional information.

Exercise 7

An official from the securities commission estimates that 25% of all investment bankers have
profited from the use of insider information. Assume that 15 investment bankers are selected
.

19. The probability that at least 10 of the selected bankers have profited from insider
information is, rounded to 4 decimals accuracy, 0.0007.

20. The expected number of investment bankers in this sample of 15, who have profited from
the use of insider information is 4.00.

An ex-banker finds the estimate of the official rather low. In fact, using his experience this ex-
banker estimates that 75% of all investment bankers have profited from the use of insider
information.

21. If the ex-banker is correct, the probability that exactly 10 of the selected bankers have
profited from insider information is, rounded to three decimals accuracy, 0.165.

Exercise 8

Phone calls at a reservation desk for a hotel arrive according to a Poisson process with a mean
of 36 calls per hour. A table with Poisson probabilities can be found on the next page.

22. It is more likely that 2 calls arrive during a 5 minute interval than that 3 calls arrive during
a 5 minute interval.

5
Table of the Poisson distribution for several values of
k
P(X k) = e- x = 0, 1, ...
k!
K = 0.5 =1 =2 =3 =6
0 0.6065 0.3679 0.1353 0.0498 0.0025
1 0.3033 0.3679 0.2707 0.1494 0.0149
2 0.0758 0.1839 0.2707 0.2240 0.0446
3 0.0126 0.0613 0.1804 0.2240 0.0892
4 0.0016 0.0153 0.0902 0.1680 0.1339
5 0.0002 0.0031 0.0361 0.1008 0.1606
6 0.0005 0.0120 0.0504 0.1606
7 0.0001 0.0034 0.0216 0.1377
8 0.0009 0.0081 0.1033
9 0.0002 0.0027 0.0688
10 0.0008 0.0413
11 0.0002 0.0225
12 0.0001 0.0113
13 0.0052
14 0.0022
15 0.0009
16 0.0003
17 0.0001

Andy, who works at the desk, decides to go out for a smoke. This will take him exactly 10
minutes.

23. The probability that no calls will be missed by Andy during his absence is 0.0025.

Exercise 9

A manufacturer must decide whether to mount, at a cost of $100,000, an advertising campaign

for a product whose sales have been rather flat. A highly successful campaign would add
$400,000 to profits, and a moderately successful campaign would add $100,000, but an
unsuccessful campaign would add nothing. Historically, 40% of similar campaigns were
highly successful, 30% moderately successful and the remainder were unsuccessful. The
company can ask a successful media consultant to assess the potential effectiveness of the
campaign. This consultant will give an advice whether or not to launch the campaign.
The for this decision problem relevant joint probability distribution is given in the table on the
next page.

6
 Positive: Negative: Do
Campaign outcome: Launch campaign not launch
Highly successful 0.36 0.04
Moderately successful 0.12 0.18
Unsuccessful 0.02 0.28

The company will make a decision concerning the launch of the campaign based on the
criterion of maximum expected monetary value (EMV). A decision tree corresponding to this
problem is provided on the next page. You may use this tree to assess the following
statements.

24. Without consultation of the media consultation, the company will not carry out the
advertising campaign.

Suppose that the media consultant gives, free of costs, an advice.

25. Even if the consultant gives a negative advice, the campaign will be launched.

26. The probability that given a positive advice from the media consultant, the outcome of the
campaign will be unsuccessful is, 0.02.

In fact, the media consultant asks a fee of $10,000 for his advice.

27. The company will not hire the consultant.

Exercise 10

Suppose that the weekly amount of time teenagers spend on the internet is normally
distributed with a standard deviation ( ) of 1.5 hours. A sample of 100 teenagers is selected at
random, and the sample mean computed as 6.5 hours.

28. The 95% confidence interval estimate of the population mean is, rounded to three
decimals accuracy, [ 6.206, 6.794].

In previous studies a researcher found the claim that the mean weekly amount of time that
teenagers spend on the internet is 7 hours. The researcher wishes to test whether this claim is
still plausible. She constructs the following hypotheses:

H0 : = 7 hours and Ha: .

29. Using the sample of 100 teenagers described earlier and a significance level = 0.05, the
researcher cannot reject the null hypothesis in favor of the alternative.

7
Figure 1: Decision tree for problems 24-27
Payoff
Highly successful

Launch campaign Moderately successful

Positive advice Unsuccessful

Do not launch

Hire consultant Highly successful

Launch campaign Moderately successful

Negative advice Unsuccessful

Do not launch

Highly successful

Launch campaign Moderately successful

Do not hire consultant Unsuccessful

Do not launch

8
Exercise 11

A production filling operation has a standard deviation ( ) of 6 gram. When in

perfect adjustment, the mean filling weight for the production process is 50 gram. We
may assume that the filling weight is normally distributed. A quality control inspector
selects at random 36 containers and uses the sample mean filling weight to see if the
process is in perfect adjustment. The sample mean filling weight is 48.6 gram. The
inspector formulates as hypotheses

H0 : 50 vs. H a : 50

30. The p-value for the corresponding z-test is, rounded to two decimals accuracy,
0.08.

31. If the true mean filling weight is 52 gram, the power of the appropriate z-test with
a significance level =0.05, is, rounded to three decimals accuracy, 0.484.

32. If the true filling weight is 52 gram, failing to reject the null hypothesis is called a
Type II error.

The plant manager wants to have a 95% confidence interval for the mean filling
weight. Moreover, the confidence interval should have a width of at most 2 grams.

33. The minimum sample size needed to construct a confidence interval satisfying the
constraints indicated by the plant manager, is 139.

Exercise 12

n, advertises that they deliver your pizza

within 15 minutes of placing an order or it is free. A sample of 25 customers is
selected at random. The average delivery time in the sample was 13 minutes and 40
seconds, with a sample standard deviation (s) of exactly 4 minutes. You may assume
that the delivery times are normally distributed. To test whether the population mean
is less than 15 minutes the following hypotheses are suggested: H 0 : 15 ,
Ha : 15

34. We can infer, at the 5% significance level, that the population mean is less than 15
minutes.

Exercise 13

Ten secretaries were selected at random from among the secretaries of a large
university. The typing speed (number of words per minute) was recorded for each
secretary on two different brands of computer keyboards. The results are summarized
in the table given on the next page.

9
 Computer Keyboard
Secretary Brand A Brand B
Amy 72 74
Betty 80 86
Carol 68 72
Donna 74 70
Ellen 84 85
Faith 75 73
Gwen 70 72
Heather 63 65
Ingrid 76 79
Jody 62 64

It is known that the typing speeds are not normally distributed. The following
hypotheses are formulated:

H 0 : The two population locations are the same.

H 1 : The two population locations differ.

To test these hypotheses, a sign test is employed.

35. These
that the brands differ with respect to typing speed.

Exercise 14

An investor is considering two types of investment. To help make her decision, she
randomly selects seven monthly returns on investment 1 and ten monthly returns on
investment 2. She finds that the sample variances of investments 1 and 2 are 225 and
118, respectively. Assume that the returns on investment 1 and investment 2 are
independent and that both are normally distributed.

36. At the 5% significance level we cannot infer that the population variances of
investment 1 and investment 2 differ.

Exercise 15

A food processor wants to compare two preservatives for their effects on retarding
spoilage. Suppose 16 cuts of fresh meat are treated with preservative A and 14 are
treated with preservative B, and the number of hours until spoilage begins is recorded
for each of the 32 cuts of meat. The results are summarized in the table below.

Preservative A Preservative B
Sample Mean 108.7 hours 98.7 hours
Sample Standard Deviation 10.5 hours 13.6 hours

10
The appropriate hypotheses to determine if the average number of hours until spoilage
begins differs for the preservatives A and B are

Ho : A B 0 vs. H a : A B 0

Assuming that the number of hours until spoilage is normally distributed and that the
population variances for the two preservatives
carries out the appropriate t-test.

37. The rejection region at .05 is: Reject if t < -2.048 or t >2.048.

38. The value of the appropriate test statistic is: t = 2.27. Hence, we conclude that at
the 5% significance level, the hours until spoilage for the two preservatives differ.

Exercise 16

A survey of 1500 Canadians reveals that 945 believe that there is too much violence
on television. In a survey of 1500 Americans, 810 believe that there is too much
television violence.

39. Using a significance level of =1%, we conclude that the proportion of Canadians
who believe that there is too much violence on television (pC) and the proportion
of Americans who believe that there is too much violence on television (pA)
differs.

40. The 99% confidence interval for the difference in the proportion of Canadians and
Americans (that is, the interval for pC pA) who believe that there is too much
violence on television is, rounded to 3 decimals accuracy: [0.044, 0.136].

The End

11