1 The marketing team of a company desires to test whether the average monthly sales is

30 lakhs (as was targeted) or less than 30 lakhs. One person from the team collects a
sample of same size 10. A second person from the team selects another sample of the
same size. The first person rejects the null if her sample mean is less than K, and the
second person rejects the null if her sample mean is less thanL. The team leader finds
thatK > L.
a) Who is testing at a lower significance level? Why?
b) Who has a higher chance of committing type II error? Why?
c) What should the person, testing with a higher significance level, do to reduce her
significance level to that of the other keeping the decision rule same? Why?
d) Suppose α =0.05 . If type I error occurs, the company spends 5 lakhs per month on
advertisement for increasing the average sales. On the other hand, if type II error
occurs with maximum probability, the company loses ₹ 100,000. Which type of error
is more important for the company? Why?

1-a If it is known that the standard deviation of monthly income of the employees of
IIMN is Rs.5000, then what is the probability that a random sample of 25 employees will
yield an average monthly income, which is within Rs.500 around the actual average
monthly income?

2 Rahul, a young IT professional, is considering the possibility of establishing an

Educational Portal for prospective MBA aspirants and wishes to conduct a survey.
Based upon the cost of setting up such a portal and the profits that may be generated,
he has arrived at the following conclusion: if there is evidence that the average
revenue per candidate is more than Rs.2500, then the portal will be established;
otherwise, the portal will not be established. Based on past experience of several other
such portals, the standard deviation of revenue is estimated to be Rs.500. Rahul wants
to be 99% certain of not committing an error of establishing the portal when the actual
average revenue is at most Rs.2500 per candidate.

a. If Rahul wishes to have a 2.5% chance of not establishing the portal when the actual
average revenue is Rs.3000, what sample size should be selected?
b. If Rahul decides that a random sample of 20 candidates should be surveyed, compute
the probability of establishing the portal when the actual average revenue is Rs.3000
per candidate.
c. Discuss the difference in your results in a) and b).
d. A sample of 25 candidates is actually surveyed yielding a sample average revenue of
Rs.2825 with a sample standard deviation of Rs.675. Construct a 95% confidence interval
for i. The population average revenue, ii. The population standard deviation of revenue.

e. Assuming the population standard deviation to be Rs.500, at 1% level of significance,

determine whether there is evidence that the population average revenue anticipated
exceeds Rs.2500.

3 An expert is asked to estimate the characteristics of the monthly disposable income of

a population. She estimates the average value to be Rs. 1000 and that 75% of the
population have a disposable income between Rs. 600 and Rs.1400.
a. What is the minimum disposable income of the top 5% of the population?
b. If the size of the population is 150000, how many people have a disposable
income below Rs. 500?

4 The maintenance department of a large IT company plans to ascertain the conditions

of PCs and the cost of their repair. There are altogether 10000 PCs, a list of which is
maintained by the company. From this list, in a pilot survey, a simple random sample
of size 100 PCs was selected without replacements. Technicians were sent out to
examine the conditions of the PCs selected, and to calculate the cost of needed
repairs. The results of this inspection were as follows: sample average repair cost =
Rs. 830; sample standard deviation of repair cost = Rs. 110. How many additional
PCs must be sampled if the estimate of the total cost of repairing all PCs (formed by
pooling the observations in the pilot and the planned sample) is to be within Rs.
100000 of the true total cost with probability 90%?
5 A proposal has an immediate investment of Rs. 600000. The estimated life of the
proposal is 6 years. The annual cash earnings are uncertain following a Normal
distribution with an expected value of annual cash earnings is Rs. 180000. The
required rate of return is 9% per annum. The annual earnings estimate of Rs. 180000
has a 30% chance of being off by more than Rs. 30000 in either direction. The
proposal will not be accepted if there is a 15% chance of losing money. Should the
proposal be accepted?
6 The ABC Company produces detergent packets of 1 kg each for the lower income
segments. Its competitor announces that it guarantees that its 1 kg packets never fall
short on weight. ABC knows that its filling machines are set at 1.10 kg and that the
standard deviation of the quantity filled is 50 grams.
a. What percentage of the packets sold by ABC is below the 1 kg level?
b. If ABC wants at most 1% to be below 1 kg, then where must it set the mean value
for the filling machine? Assume that the standard deviation does not change.
c The production manager comes up with an alternative. He says with an investment
of Rs. 60,000, they can reduce the standard deviation to 35 grams and keep the mean
setting at the same level of 1.10 kg. Assigning a cost of Rs.3 to each packet that is of
 lower weight than 1 kg (in terms of customer goodwill loss etc.), is this investment
justified? Assume that they sell a million packets a year and would want to recover
the investment in the first year of operations

7 Sketch the cumulative distribution function curve ofZ, the standard Normal
random variable.

i) Show in the graph the probability that Z is more than 1.5.

ii) Compute the probability that Z2 lies between 1 and 9. Also, show this
probability on the same graph.

8 The cumulative distribution function F (x) of a random variable X is defined as

follows
F ( x )= A -∞ < x < -1
=B -1 ≤ x < 2
=C 2≤ x <∞
If P ( X >1 )=2 /3, find the constants A,B, and C. Sketch the F ( x ).
9 Three tests have been taken by a very large number of students. LetX , score in
test 1, follow N ( 60 , σ=10 ); Y , score in test 2, follow N ( 50 , σ=8 ) ; and
Z , score ∈test 3 , follow N ( 70 , σ=9 ). The test scores are independent. Let the
combined score in the three tests be defined as W =X +3 Y + 4 Z.

a) Find the median, mode and variance of W .

b) IfX , Y and Z were not independent can you give some idea of its variance?
c) In a randomly selected group of 10 students who have taken all the three
tests, what is the probability that
i) exactly 3 of them have a combined score of more than 200?
ii) up to 2 students have a combined score of more than 200?

10 The daily return of a stock varies uniformly from -2% to 3%.What is the
expected value of the squared return? Sketch the return density.

11 Daily return of a stock is normally distributed. The return takes the value of at
least 1.78% once in 5 days and of at least 1.65% in 7 out of 10 days.
a) Sketch the above information using the graph of the return density.
b) What are the mean and standard deviation of the return?
c) Find the probability that on a given day the return will be exactly 2.5%?
 12 Suppose that the amount of time Anuja spends in a bank is exponentially
distributed with mean 10 minutes.

a) What is the probability that Anuja will spend more than 15 minutes in the
bank given that she has already spent 10 minutes in the bank?
b) What is the probability that Anuja will spend more than 20 minutes in the
bank given that she has already spent 15 minutes in the bank?
c) What is the probability that Anuja will spend more than 25 minutes in the
bank given that she has already spent 19 minutes in the bank?
d) Can you make some comment from the above three results?
e) If you pick 5 randomly selected Anuja’s visits to the bank, what is the
probability that she never waited for more than 10 minutes?

13 X is a Normal random variable with mean 21 and variance 25. If

P( X > y∨X >28)=0.625, find y.

14 Let X  be the time between two calls received at a health-care provider. It is given that the
mean time between calls is 12 minutes.
a) Find the probability that the first call will be within 5 to 10 minutes?
b) What is the probability that the time between tenth and eleventh call will exceed 2
minutes?

15 In a 5-star Hotel in Bangalore, on an average, 4 ISD calls are received every 2 hours.
It is given that on a particular day 30 ISD calls have been received.

a) What is the probability that a total of 6 ISD calls have been received between 9
p.m. to 11 p.m.?

b) What is the expected number of calls in that time interval?

16 A manufacturer produces iron rod of length L meters, where L is distributed uniformly over the
interval [15, 30]. After producing the iron rod he will send those rods to his supplier by pick-up trucks
each of which can carry exactly 20 iron rods.

Find the probability that each of 3 randomly selected pick-up trucks will contain at least 8 iron rods of
length more than 24 meters?

15. A group of students (consisting of PGPIs and PGPIIs) of one of the IIMs had
undertaken a study on Detergent Powder. The study was sponsored by Gamble and
Gamble, in a Marketing Fair, held in November, 1998. The main objective of the
study was to understand the “Brand Images” of a number of brands of Detergent
Powder. The following seven brands were included in the project:

1. Ariel Compact (AC), 2. Ariel Supersoaker (AS)

 3. Surf Excel (SE), 4. Surf with Wash Booster (SB)
5. Nirma (N), 6. Wheel (W), and 7. Rin (R).

Through several Focus Group Discussions (FGDs), the Study Team identified 19 image
parameters, which were reduced to set of eight image parameters (through a technique called
Factor Analysis). The eight image parameters thus identified are given below:
1. Innovative, 2. Indianness, 3. Gentle, 4. Young and Energetic
5. Trustworthy and Caring, 6. Versatile, 7. Powerful and efficient, and 8. Modern.
A game was designed to collect data from the female respondents only. Data on a large
number of demographic variables such as, age, education, occupation, marital status,
monthly income etc. were collected for each respondent. Each respondent was asked to
select three brands (two used by her, and one not used but she was aware of) from the set
of seven brands. Then she was asked to rate each brand on a 5-point scale (giving a rating of
l if she thought that the particular image parameter was least, and 5 if she perceived it to be
highest in the brand) on each of the eight image parameters.

Two brands, namely, Nirma and Wheel had to be discarded from analysis because of
inadequate responses on these two brands.
A measure of “Image” of a brand, as perceived by a respondent, was obtained by averaging
the ratings across all eight “Image Parameters” by the respondent for the brand. For each
respondent, one of the rated brands was then used for understanding “Image” of the brand.
The Study Group wanted to find out whether there was significant difference in “Image”
across the brands; in case there was, which brand had the “Best Image”, what score could
reasonably indicate the “Best Image” brand, and so forth. Unfortunately, the PGPIIs had
forgotten all that they had learnt in QM courses in the first year, while for PGPIs, the QM
courses had just started then. So, though, the team framed a set of questions but they
remained unanswered. Now that you have just finished QMM, can you answer following
questions with the help of the data given in Tables 1, 2 and 3?
The Project team was particularly concerned about the assumption of normality of the
ratings. Checking this assumption for each brand was considered to be time consuming and
unnecessary. The members of the team argued that if the assumption was true for one
brand, it must also be true for others. Therefore, it was decided to test the assumption for
only one of the brands.

a. Carry out a suitable test to test the normality of the Image scores using the data in Table 1
for Ariel Compact.
b. Assuming normality of the image scores for all brands, test if there is a significant difference
in the average “Image” scores among the five brands.
c. If the assumption of normality of image scores is valid, test if there is a significant
difference in the variance of “Image” scores between the first two brands.
d. If the assumption of normality is not valid, test if there is any significant difference
between the distributions of the last two brands.
e. Determine which brand has the “Best Image” on the basis of highest average image
score.
f. Calculate a 95% confidence interval for average image score for the “Best Image”
brand.
g. Assuming equality of variance, calculate a 90% confidence interval for the common
2 2
variance of image scores. ( χ 0 .95; 95 = 73.3050, χ 0 .05; 95 = 118.5232)
h. After the Marketing Fair, when the team sat down to do the data analysis, it figured
out that each respondent had given highest rating to exactly one of the three brands
she had rated. One of the members of the group argued that the choice of “Best
Image” brand must be related to the respondent’s exposure level to the
advertisement of the product. But as they did not collect any information on
respondents’ exposure to advertisement, another member suggested that the
selection of “Best Image” brand might be related to her income. Table 3 was thus
formed using the information of 212 responses. Test if this is true at 5% level of
significance.
i. Ariel Supersoaker was considered to be a competitor of Surf with Wash Booster. The
study team was thus interested in finding out if there was a significant difference in
the market share of these two brands. The first 100 respondents were asked if they
used Ariel Supersoaker; 27 of them responded in the affirmative. While out of the
next hundred, 32 respondents said they used Surf with Wash Booster. What can you
conclude (use p-value)?

Table 1: “Brand Image” Scores: For twenty randomly

Selected respondents for each brand
AC AS SE SB R
4.1 4.1 4.1 3.7 3.1
5.0 5.0 4.4 4.6 2.9
5.0 5.0 4.7 4.8 3.5
4.8 3.8 4.6 4.3 2.4
4.5 4.5 4.2 4.5 3.5
4.6 4.6 4.5 3.4 3.0
4.9 3.9 4.0 4.2 2.7
4.4 4.4 4.3 4.9 2.5
4.4 4.4 4.3 3.4 3.0
4.9 4.9 4.2 4.4 2.8
3.8 3.8 5.0 4.9 2.2
3.7 3.7 4.6 3.7 2.5
 3.5 3.5 4.9 4.6 2.5
3.8 3.8 4.9 3.7 2.2
3.9 4.4 4.5 4.0 2.0
3.6 3.6 4.9 3.9 2.6
3.7 3.7 4.9 4.2 2.7
3.6 3.6 4.9 3.3 2.1
3.8 4.8 4.8 3.8 2.3
3.2 4.2 5.0 4.0 2.2

Table 2: Summary Statistics for Table 1

AC AS SE SB R
Sum of all 20 observations 83.20 83.70 91.70 82.30 52.70
Sum of Squares of all 20 observations 352.32 355.0 422.4 343.4 142.4
7 3 9 3
SD of all 20 observations 0.5716 0.501 0.323 0.504 0.433
9 3 0 2

Table 3: Number of respondents: “Best Image” brand versus Monthly Income

Monthly Income AC AS SE SB R Total
Rs. 10000 or more 19 12 61 14 - 106
Rs. 5000 - Rs. 10000 26 22 14 10 - 72
Below Rs.5000 7 7 8 7 5 34
Total 52 41 83 31 5 212

16. Three researchers have collected household income data from three districts, A, B,
and C, independently and individually. They wish to investigate the income
distributions in these districts. The summary of their sample data are given below.

a) The first researcher takes a sample of size 9 from district A. The upper limit of
95% C.I. of the population variance of income is 29.10. The 95% C.I. of the
population mean income is (17.84, 22.16).
 b) The second researcher takes a sample from district B such that the expected value
of (n2 -1) (S22/σ 22) is 7 and the midpoint of the 95% C.I. of the population mean is
22 and length of the C.I. is 7.80.

c) The third researcher takes a sample from district C such that the expected value of
(n3 -1) (S23/σ 32) is 10 and the midpoint of the 95% C.I. of the population mean is 23.

d) The 95% C.I. of (σ 22/σ 32) is (0.20, 14.06).

From the above information (and clearly stating the assumptions which are needed)

i. Verify whether the mean income in the above three districts are equal or
not.

ii. Verify if the income variabilities in the districts AandB are equal or not.

iii. Give a 95% C.I. for the difference of mean incomes in the districts Aand
C.

18 The probability of detecting a disease in X- ray examination of a person suffering from the
disease is 3/5. The probability of diagnosing a healthy person as suffering from the disease is
1/5. 30% of the whole population is suffering from the disease. Find the probability that a
person is healthy, if after examination the person is diagnosed as suffering from the disease.

19. After the correction of quiz 1 answer books, it is found that 50% of the PGP 1 students
got more than 60% marks in DS1 paper. A random sample of twenty students is selected.

(Exact numerical answer is not required. Only steps and mathematical formulation will be
enough)

a) What is the expected number of students in that group who scored more than 60% marks?
b) What is the probability that at least 10 students in the selected sample got more than 60%
marks?
c) Can you plot a rough probability distribution of marks of that group?

20 A business office orders paper supplies from one of three vendors,V 1 ,V 2, orV 3.
Orders are to be placed on two successive days, one order per day. Thus, (V 2 ,V 3)
might denote that vendor V 2 gets the order on the first day and vendor V 3 gets the
order on the second day.

a) List the sample points in this experiment of ordering paper on two successive
days.