f
INSTITUTE OF ACTUARIES OF INDIA
EXAMINATIONS
21st May 2012
Subject CT3 Probability & Mathematical Statistics
Time allowed: Three Hours (15.00 18.00)
Total Marks: 100
INSTRUCTIONS TO THE CANDIDATES
1. Please read the instructions on the front page of answer booklet and instructions to
examinees sent along with hall ticket carefully and follow without exception
2. Mark allocations are shown in brackets.
3. Attempt all questions, beginning your answer to each question on a separate sheet.
However, answers to objective type questions could be written on the same sheet.
4. In addition to this paper you will be provided with graph paper, if required.
5. Please check if you have received complete Question Paper and no page is missing. If
so kindly get new set of Question Paper from the Invigilator
AT THE END OF THE EXAMINATION
Please return your answer book and this question paper to the supervisor separately.
IAI CT3 - 0512
Q. 1) A cricket coach records the number of runs the players in his squad scored in a tournament.
Each player got a chance to bat at least once. He presents the data in a stem and leaf diagram:
KEY: 2 | 7 means 27 runs scored
0 1 1 2 7
1 2 5 5
2 3 7
3 6
4 0
5 0 9
a) What is the range of the data? (1)
b) What is the median number of runs scored? (1)
c) Compute the mean number of runs scored. (2)
[4]
Q. 2) The observed mean (and standard deviation) of the number of claims and the individual losses
over a given period are 7.6 (3.2) and 197,742 (52,414), respectively. Assume the variable for
number of losses is independent of the variable for individual loss sizes. Determine the mean
and standard deviation of aggregate claims.
[5]
Q. 3) The random variables are independent and normally distributed with mean 0 and
variance 1. Let be an unknown parameter.
Suppose you are given observations and such that:
a) Write down the regression model. (1)
b) Derive the expression for the least square estimator of . (3)
You are given that the values of and are 0.6 and 1.8 respectively.
c) Calculate the value of the estimator (1)
d) Suppose now . Find and hence provide a 95% prediction
interval for (7)
[12]
Q. 4) Based on a Normal random sample of size 100, a 90% confidence interval for the population
mean turned out to be (20, 40). Find a 95% confidence interval for the population mean based
on this information. [5]
Page 2 of 6
IAI CT3 - 0512
Q. 5) Let N be a Poisson random variable with mean .
Define a random variable Y: where , > 0 are given constants.
An actuarial student decided to construct a random variable X based on Y by playing the
following game:
He tosses an unbiased coin
If it turns up Head, he assigns X = Y
If it turns up Tail, he assigns X = 0
a) Show that the moment generating function of X can be expressed in terms of the
moment generating function of Y as below:
(2)
b) Hence or otherwise show that the probability distribution of X can be expressed as:
Value Probability
(7)
c) Consider two such independent random variables X1 and X2 (constructed similar to X
above) with the following values of the parameters:
Variables
X1 1 2 1.0
X2 2 3 1.5
Compute . (8)
[Hint: ]
[17]
Q. 6) A university runs a 3-year B.Sc degree course in Statistics. The course is divided over 6
semesters each consisting of 5 credit papers over the three year period. Each credit paper is
assessed on a maximum possible 100 marks and is recorded as integers.
At the end of the course, the university ranks the students based on a measure called grade
point average which is the average of marks obtained over all credit papers examined over 3
years.
Assume that in each subject, the instructor makes an error of quantum k in awarding marks with
probability where Assume that these errors occur independently.
a) Show that the probability of no error is . (2)
Page 3 of 6
IAI CT3 - 0512
b) State the approximate distribution of quantum of error in a given students final grade
point average using the Central Limit Theorem. (3)
c) Hence show that there is only a 17.7% chance that his final grade point average is accurate
to within 0.05. (3)
[8]
Q. 7) Suppose that are independent and identically distributed Poisson random
variables.
a) Find the maximum likelihood estimator of . (4)
b) Suppose that rather than observing the random variables precisely, only the events
for i =1, 2 n are observed. Find the maximum likelihood
estimator of under the new observation scheme.
(6)
[10]
Q. 8) Let and constitute a random sample of size 2 from the population.
For testing H0: = 0 versus H1: > 0, we have two competing tests:
a) Find the value of C so that Test 2 has the same P (Type I error) as that of Test 1. (2)
b) Compute the P (Type II error) of each test for a given value of = 1> 0. (3)
c) Comment on your results as obtained in part (b)? (1)
[6]
Q. 9) Suppose 2,000 finished products both from Factory A and Factory B were chosen at random by
the CEO of the Company and verified if the same have any defects. Following are the results:
Factory A Factory B
Non-defective 1,816 1,986
Defective 184 14
Perform a chi-square test on this contingency table to show that there is overwhelming evidence
against the hypothesis that there is no association between the factory and whether or not the
product is defective. State your level of significance.
[5]
Page 4 of 6
IAI CT3 - 0512
Q. 10) An actuarial student fits the following linear regression model to a given data:
Here is are independent, identically distributed random variables, each with a normal
distribution with mean 0 and unknown variance2.
The following information is available:
n=7
95% confidence interval for
Calculate what portion of the total variability of the responses is explained by the model.
[5]
Q. 11) To measure the effect of a fitness campaign, a gym instructor devised two types of sampling
design:
Design 1: Here he randomly sampled 5 members before the campaign and measured their
weights (X1), and another 5 after the campaign (X2). The results (along with some summary
statistics) are as follows:
Weights Xk Xk *Xk X1 *X2
Before: X 1 168 195 155 183 169 870 152,324
148,265
After: X2 183 177 148 162 180 850 145,366
Design 2: Here he decided to measure the weights of the same people after(X3), as before the
campaign. The results (along with some summary statistics) are as follows:
Weights Xk Xk *Xk X1 *X3
Before: X1 168 195 155 183 169 870 152,324
149,032
After: X3 160 197 150 180 163 850 145,878
a) Is it appropriate to assume that under each of the two respective sampling design schemes
the two samples of data obtained constitute two independent random samples? Explain. (1)
b) Calculate a 95% confidence interval for the mean weight loss during the campaign on the
basis of results obtained under each of the two respective sampling design schemes. (8)
c) What can you conclude about the effectiveness of the campaign from the two confidence
intervals obtained in part (b)? Comment on the relative width of the two intervals as well? (2)
[11]
Page 5 of 6
IAI CT3 - 0512
Q. 12) Many businesses have music piped into the work areas to improve the environment. At a
company an experiment is performed to compare different types of music. Three types of
music country, rock and classical are tried, each on four randomly selected days. Each day
the productivity, measured by the number of items produced, is recorded. The results appear
below:
Music Type Productivity (y) y y2
Country 857 801 795 842 3,295 2,717,039
Rock 791 753 781 776 3,101 2,404,827
Classical 824 847 881 865 3,417 2,920,771
9,813 8,042,637
[Draw all your statistical inferences at 5% significance level]
a) Perform an analysis of variance to show that the mean number of items produced differs
for at least two of the three types of music. (5)
b) Show that the mean number of items produced in rock music is significantly worse than
those produced in the other two. (4)
c) Show that it is statistically difficult to ascertain which music has the best effect in terms of (3)
the mean number of items produced.
[12]
******************************************
Page 6 of 6
