Copyright reserved

UNIVERSITY OF PRETORIA
FACULTY OF NATURAL AND AGRICULTURAL SCIENCES
DEPARTMENT OF STATISTICS
MATHEMATICAL STATISTICS 111
EXAMINATION: 28 MAY 2010
AVAILABLE TIME: 3 HOURS MARKS: 100

External Examiner: Prof C F Smit

Internal Examiners: Mrs. EM Basson
Mrs. HF Strydom

SURNAME AND INITIALS

STUDENT NUMBER

SIGNATURE

 Answer all questions in the spaces provided.

 Write down the formulae you use.
 Marks will be awarded for notation. Show all steps.
 Give answers correctly up to 3 decimal places where necessary.
 A list of formulae is attached.
 Cell phones may not be used and must be switched off.

QUESTION TOTAL MARKS AWARDED

1 6
2 9
3 11
4 20
5 7
6 13
7 6
8 9
9 8
10 11
100
QUESTION 1
1  n 2 1 n  
2
1 n
  yi  y   n  1  yi  n   yi   .
2 2
(a) Show that s 
n  1 i 1 i 1  i 1  

[4]

3
(b) Show that the sum of the deviations of a set of measurements about their mean is equal to
n
zero, that is,  y
i 1
i  y  0 .

[2]
Total for question 1: 6
QUESTION 2

Consider two events A and B that are not mutually exclusive.

(a) Use union and intersection of the events A and B as well as complimentary events of A and B
to:

(i) express A  B in terms of two mutually exclusive events.

(2)
(ii) express B in terms of two mutually exclusive events. (You may not use  ).

(2)
[4]

4
(b) Suppose A and B are statistically independent events.

(i) Show by means of the rules of probability that

P( A  B)  1  P( A )  P( B ) .

(3)
3 3
(ii) Given that P ( A )  and P ( A  B )  ,
4 5
calculate P ( A  B ) and P ( B ) .

(2)
[5]
Total for question 2: 9

5
QUESTION 3

(a) A firm places three orders for supplies among five different distributors. Each order is
randomly assigned to one of the distributors, and a distributor may receive multiple orders.
Use counting rules to find the probabilities of the following events:

(i) All orders go to different distributors.

(2)
(ii) All orders go to the same distributor.

(2)
(iii) Exactly two of the three orders go to the same distributor.

(2)
[6]

6
(b) Jim has applied for a bursary for next year. His estimates of the probabilities of getting a
certain final mark and the probabilities of getting the bursary given the final mark, are given in
the table:

Final Mark 80-100% 70-79% 60-69% 50-59% <50%

(A) (B) (C) (D) (Fail)

P(getting final mark) 0.20 0.15 0.50 0.10 0.05

P(getting bursary | final mark) 0.90 0.75 0.40 0.15 0.00

You subsequently hear that he was awarded the bursary. In other words, given that he was
awarded the bursary, what is the probability:

(i) that he obtained a final mark above 70%?

(4)
(ii) that he failed?

(1)
[5]
Total for question 3: 11

7
QUESTION 4

(a) For each question below, encircle the correct answer:

(i) Trials are dependent for a random variable with the following distribution:

Binomial Geometric Hypergeometric Poisson Normal

(ii) When the sample size n is large and the probability of success p is small ( np <7,
roughly), binomial probabilities can be approximated by probabilities from the
following distribution:

Geometric Poisson Hypergeometric Uniform Exponential

(iii) Which distribution may serve as a model for the intervals of time between break
-downs of a computer?

Binomial Geometric Hypergeometric Uniform Exponential

(iv) Which distribution may serve as a model for the number of times a total of 6 occurs
when 2 dice are thrown 5 times?

Binomial Geometric Hypergeometric Uniform Exponential

(v) Which distribution may serve as a model for the precise masses of packets of 36
biscuits of the same brand and weight?

Binomial Poisson Uniform Normal Exponential

(vi) Which distribution may serve as a model for the number of telephone calls received
each day by a telephone counseling service?

Binomial Geometric Hypergeometric Poisson Exponential

et
(vii) The distribution of the random variable with moment-generating function is:
2  et

Binomial Geometric Hypergeometric Poisson Uniform

[7]

8
 2
(b) Let Y be a discrete variable with mean µ and variance σ and a and b constant values.

(i) Use the probability function p ( y ) to show that E ( aY  b)  a  b .

(2)
(ii) Use the definition of the variance of a random variable and the properties of
expected values to show that V ( aY  b)   2V (Y ) .

(2)
(iii) Derive the relation between the second-order central moment and second order
moment about the origin.

(2)
(iv) Derive the relation between the second-order moment about the origin and the
factorial moments.

(2)
[8]

9
(c) Let Y have a binomial distribution with parameters n and p. Derive an expression for the
variance of Y . Hint: Use the result derived in (b)(iv). You may assume that E(Y) = np.

[5]
Total for question 4: 20

10
QUESTION 5

A company has 10 personal computers (PCs) in its warehouse. Although all are new and still in
boxes, four do not currently function properly. One of the company’s offices requests 5 PCs and
the warehouse foreman randomly selects 5 from the stock of 10 and ships them to the requesting
office.

(a) Calculate the probability that all five of the PCs are not defective.

[2]
(b) The office that requested the PCs returns the defective ones for repair. If it costs R300 to
repair each PC, calculate the mean and the variance of the total repair cost. (It is not
neccessary to derive the appropriate expressions).

[3]
(c) According to Tchebysheff, the probability is at least 0.75 that the total repair cost lies in the
interval (110.102, 1089.898). Interpret this interval in terms of standard deviations.

[2]
Total for question 5: 7
11
QUESTION 6

(a) Let Y have a uniform distribution with probability density function

 1
 , 1  y   2 .
f ( y )   2  1
 0, otherwise

.
(i) Derive the mean of Y .

(2)
(ii) Derive the distribution function of Y.

(2)

12
 (iii) Use (ii) to derive the median of Y.

(2)
(iv) What is the relationship between the mean of Y and its median in this case?
Why does this relation exist?

(1)
[7]

13
(b) Suppose a parachutist lands at a random point on a line between markers A and B. Calculate
the probability that she is closer to A than to B.

[3]
(c) Suppose three parachutists, operating independently, each lands at a random point on a line
between markers A and B. What is the probability that exactly one of the three lands closer to
A than to B?

[3]
Total for question 6: 13

14
QUESTION 7

The length of time Y necessary to complete a key operation in the construction of houses has an
exponential distribution with mean 10 hours.
 0, y0
(a) The distribution function is given by F ( y )    y / 10
. Use F ( y) to calculate the
1  e , y0
probability that the length of time to complete the operation is more than 12 hours, given that
the time used to complete the operation is already more than 10 hours.

[3]
2
(b) The formula C = 100 + 40Y + 3Y relates the cost C of completing this operation to the time to
completion of the operation. Calculate the mean of C.

[3]
Total for question 7: 6
15
QUESTION 8

Suppose Z is a random variable with moment-generating function mZ (t ) .

(a) Let Y  aZ  b , with a and b constants. Show that the moment-generating function of Y is
given by mY t   e tb mZ at  .

[3]
t2 / 2
(b) If Z ~n(0,1), then M Z (t )  e . Let Y ~n(μ,σ2). Use the result in (a) to determine the
moment-generating function of Y .

[3]

16
(c) Demonstrate how the moment-generating function of Z generates the moments of this
standard normal distribution.

[3]
Total for question 8: 9

QUESTION 9

Men’s shirt sizes are determined by their neck sizes. Suppose that adult men’s neck sizes are
approximately normally distributed with a mean of 40 cm and standard deviation of 2 cm. A retailer
sells men’s shirts in sizes S, M, L, XL where the shirt sizes are defined in the following table:

Shirt Size Neck Size (in cm)

S <37
M 37-40.5
L 40.5-44
XL >44

(a) What proportion of shirts of size S should be manufactured?

[2]

17
(b) What proportion of shirts of size M should be manufactured?

[2]

(c) If you wanted to define categories S, M, L, XL so that each category contained 25% of the
total population of adult males, what neck sizes must you assign to each of these
categories?

[4]
Total for question 9: 8

18
QUESTION 10

(a) Contracts for two construction jobs are randomly assigned to one or more of three firms A, B
and C. Let Y1 denote the number of contracts assigned to firm B and Y2 the number of
contracts assigned to firm C. Each firm can receive 0, 1 or 2 contracts.

(i) Determine the joint probability function for Y1 and Y2.

(3)
(ii) Determine and name the marginal distribution of Y1.

(2)
[5]

19
(b) The joint distribution of two variables X and Y is given by the entries in the table below:

-1 0 1

Y -1 1/6 1/3 1/6

1 1/6 0 1/6

(i) Determine F(0,1).

(2)
(ii) Calculate the covariance of X and Y.

(3)
(i) Are X and Y independent? Why?

(1)
[6]
Total for question 10: 11
TOTAL: 100

20