CSIR NET-Mathematical Science

2014-2020 Statistics [PYQs]

This paper are contains previous year questions of CSIR-NET
Mathematical Sciences [Statistics].

PANKAJ KUMAR

Visit us: http://pstatisticstutorials.com/

Youtube: PStatisticsTutorials
Telegram Channel: https://t.me/Pstatisticstutorials
S.No. CSIR Previous Year Question Paper Page No.
1. CSIR NET Nov-2020 1-16
2. CSIR NET Dec-2019 17-23
3. CSIR NET June-2019 24-30
4. CSIR NET Dec-2018 31-37
5. CSIR NET June-2018 38-44
6. CSIR NET Dec-2017 45-51
7. CSIR NET June-2017 52-58
8. CSIR NET Dec-2016 59-65
9. CSIR NET June-2016 66-72
10. CSIR NET Dec-2015 73-79
11. CSIR NET June-2015 80-86
12. CSIR NET Dec-2014 87-93

Visit us: http://pstatisticstutorials.com/

Youtube: PStatisticsTutorials
Telegram Channel: https://t.me/Pstatisticstutorials

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page1

CSIR NET Nov-2020
Single Choice Questions [SCQ]
1. In an examination involving multiple choice questions , a student works out the solution in
50% of the questions .In the remaining questions the student guesses the answer . However,
when the answer is guessed the probability that it is correct is 0.30 .When the student works
out the solutions it may be wrong with probability 0.10. If the answer to a particular question
is correct, what is the probability that the student guessed the answer?
(a) 0.25
(b) 0.50
(c) 0.90
(d) 0.30
Ans: a
Explanation:
Let Student works out the solution in 50% of the questions
Remaining questions the student guesses the answer
Student gives the answer by solution or gauss
( ) ; ( )
( | ) ; ( | )
( ) ( | )
( | ) ( ) ( ) ( ) (
| |)
Hence option (a) is correct.
2. Let , , …be i.i.d. random variables having a - distribution with 5 degrees of freedom.
Let be constant. Then the limiting distribution of ( ) is
√
(a) Gamma distribution for an appropriate value of
(b) -distribution for an appropriate value of
(c) Standard normal distribution for an appropriate value of
(d) A degenerate distribution for an appropriate value of
Ans: c
Explanation:
Since , , …be i.i.d. random variables having a - distribution with 5 degrees then
( ) and ( )
We know that sum of independent of chi-square variate is also an chi-square variate with
degree of freedom of their sum of degree of freedom. Therefore ;
( ) and ( )
Using Central Limit Theorem
( )
√
Thus ( ) ( ) if we take
√ √ √
Hence option (c) is correct.
3. Consider a Markov chain , , , …with state space . Suppose are two states
which communicate with each other. Which of the following statements is NOT correct?
(a) Period of = period of
(b) is recurrent if and only if is recurrent
(c) ( | )= ( | ) for all k

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page2

 (d) ( | )= ( | )
Ans: c
Explanation:
Given Markov chain , , , …with state space . If are two states which
communicate with each other.Then their corresponding periodicity will be same but
converse not always true and also if state recurrect then state is also recurrent
converse also true.
( | ) ( | ) for all k
Because state k may are may not be communicate with state
( | )= ( | )
Because communicate with each other.
Hence option (c) is correct.
4. Suppose that X has uniform distribution on the interval[ ]. Let Y denote the greatest
integer smaller than or equal to X. Which of the following is true?
(a) P( )=
(b) P( )=
(c) E( )=50
(d) E( )=
Ans: b
Explanation:
Since [ ] ( )
Y denote the greatest integer smaller than or equal to X. Then [ ]
P( ) ([ ] ) ( ( )) ∫ ( ) ∫
( )
E( ) ([ ]) ∫ ( ) ∫
Hence option (b) is correct.
5. Let , …, be i.i.d. random variables with common pdf
( )
𝒇( | )=
for where is an unknown parameter. Then the statistic T=∑ is
(a) Sufficient, but not complete
(b) Sufficient, but not minimal sufficient
(c) Complete sufficient
(d) Neither complete nor sufficient
Ans: c
Explanation:
Let , …, be i.i.d. random variables with common pdf
( )
𝒇( | )=
( )
( ) ∏ 𝒇( ) [ ] ∑ ( ) ( )
Using Neymann-Factorization Theorem
T=∑ is sufficient as well as minimal sufficient and also complete for
Hence option (c) is correct.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page3

6. Consider the pdf given by
( )
𝒇( | ) = ( )]
,
[
Based on one observation X with the above pdf, a UMP test of size for testing
: versus is
(a) X k for some k such that [ ]
(b) X k for some k such that [ ]
(c) X k for some k such that [ ]
(d) X k for some k such that [ ]
Ans: a
Explanation:
( )
𝒇( | ) = ( )]
,
[
: versus
( )

[ ( )] ( )
( ) [ ( ) ] √ ( )

[ ( )]

, where 𝒇( ) and [ ]
Hence option (a) is correct.
7. Consider 35 i.i.d. observations , …, and , … , . Let R be the Wilcoxon’s
rank sum statistic based on the ranks of the X’s in the combined sample. Then the expected
value of R is
(a) 270
(b) 300
(c) 360.5
(d) 330.5
Ans: a
Explanation:
Let N i.i.d. observations , …, and , … , . Let R be the Wilcoxon’s rank
sum statistic based on the ranks of the X’s in the combined sample. Then
( ) ( )
( ) and ( )
and given that
Therefore
( ) ( )
( )
( ) ( )
( )
Hence option (a) is correct.
8. Let Consider two –way ANOVA ,where the observations satisfy the linear model
= + + ,1
( )= 0, Var ( )= , ∑ =∑ =0
In the set-up
(a) is estimable

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page4

 (b) is estimable
(c) is estimable
(d) is estimable
Ans: c
Explanation:
Given two-way ANOVA model
= + + ,1
with the following assumption ( )= 0, Var ( )= ,∑ =∑ =0
∑ ∑ ∑ (1)
∑ ∑ ∑ (2)
Subtracting (2) from (1), we get
∑ ∑ ( ) ∑ ∑ (3)
Taking expectation on both side, we get
(∑ ∑ ) ( ) (∑ ∑ )
( ) [ (∑ ∑ )]
Therefore ( ) is estimable and there estimate (∑ ∑ )
Hence option (c) is correct.
9. Let and be two i.i.d. multivariate normal random vectors with mean and
positive definite dispersion matrix ∑ Then which of the following random variables always
has a central chi- square distribution
(a) ( ) ( )
(b) 2( ) ( )
(c) 2( ) ∑ ( )
(d) ( ) ∑ ( )
Ans: d
Explanation:
Result: ( ∑) then ( )∑ ( )
( ) and
( ) ( ) ( ) ( ) ( ) ( ) ∑
[ and be two i.i.d. multivariate normal random vectors with mean and
positive definite dispersion matrix ∑]
(( ) ) ( ∑) (( ) ) ( ) ∑ ( )
Hence option (d) is correct.
10. 10 units are chosen by simple random sampling without replacement from a population of
size 100. Consider the sample variance ∑ ( ̅) = .An unbiased estimate of
population variance = ∑ ( ̅ ) is
(a)
(b)

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page5

 (c)
(d)
Ans: *Incorrect option.
Explanation:
SRSWOR ( ) ( )
= ∑ ( ̅ ) and = ∑ ( ̅) ( )
∑ ( ̅) ( ) ( )
( ) ( )
is unbiased estimate of
11. Consider a randomized block Design with b block and k treatments. Let the observation
corresponding to the treatment and the block be ,1 which
satisfies the usual linear model. Which of the following is true?
(a) The estimates of any two treatment contrasts are uncorrelated
(b) The error sum of squares has degrees of freedom
(c) The estimate of any treatment contrast is uncorrelated with the estimate of any block
contrast
(d) The correlation between the estimates of two treatment contrasts is always negative
Ans: c
Explanation:
Randomized Block Design with one observation per cell, and data satisfying the standard
linear model

Where, ( ) and ( )
General Mean effect
Mean treatment effect
Mean block effect
The estimate of the parameters , and are
1. ̂ ̅ ( ̂)
2. ̂ ̅ ̅ ( ̂)
3. ̂ ̅ ̅ (̂)
Treatment contrast: ̂ ̂ ̅ ̅
Block contrast: ̂ ̂ ̅ ̅
( ̂ ̂ ̂ ̂ ) (̅ ̅ ̅ ̅ )
(a) The estimate of any treatment contrast is uncorrelated with the estimate of any block
contrast.
(b) The estimates of any two treatment contrasts are may or may not be uncorrelated.
(c) The error sum of squares has ( )( ) degrees of freedom
(d) The correlation between the estimates of two treatment contrasts is not always
negative
Hence option (c) is correct.
12. The maximum and the minimum values of 5x +7y , when |x|+|y| 1 are

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page6

 (a)5 and -5
(b)5 and -7
(c)7 and -5
(d)7 and -7
Ans: d
Explanation:
Let ( ) and |x|+|y| 1

Solving above equation we get the point ( ) ( ) ( ) ( )

Thus the maximum and minimum value of ( ) are 7 and -7 respectively.
Hence option (d) is correct.
Multiple Choice Questions [MCQ]
13. Let X and Y be independent random variables with E(X) = E(Y)=0 and Var(X)= Var(Y)=1
Let Z = X+Y. Which of the following statements are correct?
(a) P(|Z| )
(b) E(|Z|) √
(c) E( )=2
(d) P(Z ) (Z )
Ans: a, b, c
Explanation:
Using Chebychev’s Inequality
Let X be a random variable with ( ) and ( )
( ) ( )
| | or | |
Let ( ) ( )
For option (a)
( )
P(| | ) P(| | )
For option (b)
Using convex function condition on Expectation
E(|Z|)= (√ ) √ ( ) √ (| |) √
For option (c)
( ) ( ) ( ) ( )
For option (d)
From the given information we can’t determine the probability because not known the
information about the distribution.
Hence option (a), (b) and (c) are correct.
14. For n let , …, be random variables such that E( )=0 and E( )=1 for all and
and E ( )= for all j .Which of the following statements are true?
(a) if and only if , ··· are independent
(b) ( + +···+ )= n if and only if , …, are independent
(c) ( ) if and only if , …, are pairwise independent.
(d) ( ) if and only if 0
Ans: d
Explanation:

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page7

 Let , …, be random variables such that E( )=0 and E( )=1 for all and and
E( )= for all
For option (a)
If , then , …, are not always independent.
For option (b) and (c)
( + +···+ )= n, then ( ) and are not always independent.
For option (d)
If ( ) , then ( ) ( ) ( ) ( )
If ( ) then
( ) ∑ ( ) ∑ ( )
Hence option (d) is correct and option (a) ,(b) and (c) are not correct.
15. Consider a Markov chain with a countable state space . Identify the correct statements.
(a) If the Markov chain is aperiodic and irreducible then there exists a stationary distribution
(b) If the Markov chain is aperiodic and irreducible then there is at most one stationary
distribution
(c) If is finite then there exists a stationary distribution
(d) If is finite then there is exactly one stationary distribution
Ans: b, c
Explanation:
For option (a)
If the Markov chain is aperiodic and irreducible then the Markov chain has always unique
stationary
For option (b)
If the Markov chain is aperiodic and irreducible then there is at most one stationary
distribution.
For option (c)
If is finite then there exists a stationary distribution
For option (d)
If is finite then Markov chain has unique or infinitely many stationary distribution
because it may reducible or irreducible.
Hence option (b) and (c) are correct.
16. Consider a Markov chain with transition probability matrix

( )
Let = ( )
Then which of the following statements are correct?
(a) is a stationary distribution
(b) If is a stationary distribution , then =

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page8

 (c) The Markov chain is periodic
(d) The Markov chain is irreducible
Ans: a
Explanation:
For option (a) and (d)
Since given Markov chain with transition probability matrix

( )
Reducible and double stochastic Markov chain (each rows and column sum are equal to
one), then it has infinitely many stationary distribution and one of these stationary
distribution is ( ) ( ) where
For option (b)
If is a stationary distribution then it may or may not be equal to because given
Markov chain is reducible.
For option (c)
State 2 and 3 are aperiodic because and , then Markov chain is
periodic or aperiodic.
Hence option (d) is correct and option (a), (b) and (c) are not correct.
17. Suppose , ,..., , are i.i.d. random variables with characteristic function
( ) E[ ] where is the parameter of the distribution . Let Z= + +···
. Then for which of the following distributions of would the characteristic function of
Z be of the from (t; ) for some ?
(a) Negative Binomial
(b) Geometric
(c) Hypergeometric
(d) Discrete uniform
Ans: a
Explanation:
Suppose , ,..., , are i.i.d. random variables with characteristic function
( ) E[ ] where . Let Z= + +···
( )
[ ] [ ] [ ] [ ] [ ] ( )

This will be hold only for Negative Binomial distribution when it have i.i.d random
variable or corresponding probability will be same.
Hence option (a) is correct.
18. Let , ,…, be i.i.d. with the common pdf 𝒇(x| )= , for x where is an
unknown parameter .which of the following estimators of are consistent?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page9

 (a) ∑
(b) ∑ ( )
(c) ∑
(d) ∑ ( )
Ans: d
Explanation:
Let , ,…, be i.i.d. with the common pdf 𝒇(x| )= , for x where is an
unknown parameter.
( ) ∫ ( ) ∫ ∫
Therefore ∑ ( ) convergent in probability {By Khinchine’s Theorem}
Now, ∑ ( ) ∑ ( )
We know that convergence in probability implies consistency
Thus ∑ is consistent for
( )
Hence option (d) is correct.
19. Let , ,…, be i.i.d. with the common probability mass function ( | )
( ) ,x=0 or 1 , and 0 . Then
(a) The method of moments estimator of is ∑
(b) The MLE of is
(c) The method of moments estimator of is
(d) The MLE of is min{ ∑ }
Ans: d
Explanation:
Let , ,…, be i.i.d. with the common probability mass function ( | )
( ) ,x=0 or 1 , and 0
Likelihood function
∑
( ) ∏ ( | ) ( ) ∑
∑ ∑
( ) ̂ ∑
Since MLE of will be ∑ when
But , then ̂ { ∑ } because ̂ .
Similarly, ̂ ∑ when .
But Then method of moment’s estimator of is { ∑ } because
̂ .
Hence option (d) is correct.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page10

20. Let the pdf of X be ( | ) for 0 where is unknown parameter. Which
of the following are 100(1 ) confidence intervals for ?
(a) [ ]
√
(b) [ ]
√ √
(c) [ ]
√ √
(d) [ ]
Ans: a, c
Explanation:
( | )= for 0 where
Let
𝒇( ) 𝒇( | ) | |
For option (a)
( ) ( ) (√ ) (√ )
√ √

∫ 𝒇( ) ∫ [ ]
√ √ √
For option (b)
( ) ( ) ( ) ( )

∫ 𝒇( ) ∫ [ ]
For option (c)
√ √ √ √ √ √
( ) ( ) ( )
√ √ √ √ √ √
√ √ √
√ √ √ √ √
( ) ∫√ 𝒇( ) ∫√ [ ]√
√ √
√ √ √
For option (d)
( ) ( ) ( ) ( )
∫ 𝒇( )
Hence option (a) and (c) are correct.
21. X has binomial distribution with parameters and p. Suppose that n is given and the
unknown parameter p has prior distribution, which is uniform on the interval [0, 1]. Consider
the squared error loss function and the observation X=n. Identify the correct statement.
(a) The Bayes estimate of p is ( )
⁄( )
(b) The Bayes estimate of p is
⁄( )
(c) The median of the posterior distribution of p is
(d) The median of the posterior distribution of p is ( )
Ans: a, c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page11

Explanation:
( ) 𝒇( ) ( ) ( )
( ) ( )
( ) 𝒇( ) ( ) ( ) ( )
( ) 𝒇( ) ( ) ( ) ( )
( ) ∫
Posterior probability
( )
( | ) ( )
( )
Bayes estimate of p under squared error loss function is
̂ ( | ) ∫ ( )
Thus Bayes estimate of p is ( )

∫ ( | ) ( )
⁄( )
Thus median of the posterior distribution of p is
Hence option (a) and (c) are correct.
22. Let , , be a random sample from the uniform distribution on the interval (0, ).
Suppose the prior distribution of is uniform on the interval (0,1). Let ( ) = max
[ , , ].Consider the squared error loss function .Which of the following statements are
necessarily true?
(a) Bayes estimator of is unique
(b) is a Bayes estimator of
( )
(c) ( ) is a Bayes estimator of
( )
(d) is a Bayes estimator of
( )
Ans: a, d *(Incorrect)
Explanation:
( ) ( ) ( ) ∫ ( )
( ) ( )
Let ( ) = max [ , , ]
( ) [ ( )] ( )
( ) ( | ) ( )
( ) ∫ ( )
Posterior probability
( )
( | ) ( ) ( )
Under the squared error loss function Bayes estimate
̂ ( ) ( | ) ∫ ( | ) ( )
( )

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page12

 [ ( )= max , , ]
̂ ( ) ( )
( )
Hence option (a) is correct.
23. Consider the Gauss- Markov model = + , where ( ) and Dispersion
( )= . Suppose that p Which of the following are correct?
(a) Least- squares estimate of is unique
(b) Least- squares estimate of an estimable linear function of is unique
(c) Least- squares estimate of X is unique
(d) Determinant ( X) 0
Ans: b
Explanation:
Gauss-Markov model = + , where ( ) and Dispersion
( )= . Suppose that p
If ( ) then ( )
But Rank( ) is not mention, then least-square estimate of is not unique.
If estimable then their least-square estimate is unique.
| |
Hence option (b) is correct.
24. Let p and 1 . Consider a multiple linear regression problem with p independent
variables , , and a dependent variable Y. Suppose that the correlation between Y
and is and the correlation between and is also for all 1 . Which of the
following are correct?
(a) The multiple correlation between Y and ( ,…, ) is larger than or equal to
(b) The multiple correlation between Y and ( ,…, ) will be if
(c) The multiple correlation between Y and ( ,…, ) will be only if =0
(d) The multiple correlation between Y and ( ,…, ) tends to 1 as p
Ans: a
Explanation:
( ) and ( )

; | | and | |

( ) ( )
Since
If then
Hence option (a) is correct.
25. Let n and 0 be fixed. Let , be i.i.d. normal random variables with mean
zero and variance For =1,…, n define
and = . Further, let ( , ,…, ), and
=( , , , , , , ).
Which of the following statements are correct?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page13

 (a) has a multivariate normal distribution
(b) There exists a constant C, such that has a chi-square distribution
(c) has a multivariate normal distribution
(d) E( )
Ans: a, b, c, d
Explanation:
( )
( ); ( )
( ) ( )

=( , ,…, ),

[ ]
Since symmetric and positive definite matrix.
has multivariate normal distribution.
There exists a constant C ;
The value of C is depend on if we take then it will hold the condition.
[ ] ; [ ]
Therefore has multivariate Normal distribution.
E( ) ( )
; ( )
Hence option (a), (b), (c) and (d) are correct.
26. For circular systematic, which of the following are correct?
(a) Sample mean is an unbiased estimate for population mean but sample variance is not an
unbiased estimate for population variance
(b) Sample mean and sample variance are unbiased estimates for population mean and
population variance respectively
(c) Sample mean is not an unbiased estimate for population mean but sample variance is an
unbiased estimate for population variance
(d) Neither sample mean nor sample variance is an unbiased estimate for their population
counterparts
Ans: a
Explanation:
( ̅) ∑ ( ∑ ) ∑ (∑ ) ̅
where r stands for the sample selected with the random start r
( ̅) ∑ (̅ ̅)
( ( ̅)) ( ∑ (̅ ̅) )
Sample mean is an unbiased estimate for population mean but sample variance is not an
unbiased estimate for population variance.
Hence option (a) is correct.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page14

27. In a Randomized Block Design with one observation per cell, and data satisfying the
standard linear model, which of the following are correct?
(a) Mean treatment effects are estimable
(b) Mean block effects are estimable
(c) Treatment-Block interactions are NOT estimable
(d) Treatment and block effects as well as treatment-block interactions are estimable
Ans: a, b, c
Explanation:
Randomized Block Design with one observation per cell, and data satisfying the standard
linear model

Where, ( ) and ( )
General Mean effect
Mean treatment effect
Mean block effect
The estimate of the parameters, and are
1. ̂ ̅ ( ̂)
2. ̂ ̅ ̅ ( ̂)
3. ̂ ̅ ̅ (̂)
There is no treatment-block an interaction exists in R.B.D, then it will not be hold
estimability criteria.
Hence option (a), (b) and (c) are correct.
28. Suppose ( ) for is a continuous hazard function of a non-negative random variable X,
where ( ) . Which of the following statements are always true?
(a) ( ) is a hazard function
(b) ( ) is also a hazard function
(c) ( ) for c 0 is also a hazard function
(d) ( ) is a hazard function
Ans: b, c
29. Let = where are i.i.d with p.d.f
( ) | |, Let
∑ and = ([ ] ) , the sample percentile. Which of the following are
correct?
(a) is a consistent and asymptotically normal estimator of
(b) is a consistent asymptotically normal estimator of
√
(c) The asymptotic variance of is
(d) The asymptotic variance of is
Ans: a, b, c, d
Explanation:
Let = where are i.i.d .with pdf
( ) | |, Let

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page15

 ∑ and = ([ ] )
, the sample percentile
( ) ( ) ( )
( ) ( ) ( )
Using CLT
√ ( ) ( )
√

( ) & ( )
Therefore
(a) is a consistent and asymptotically normal estimator of
(b) The asymptotic variance of is
Asymptotically distribution of sample quartile
Let ( ) with cdf F(x) and continuous density ( ). If ( ) ,
then
( )
√ ( ̂ ) ( ( )
)
̂ Sample Quantile ; Population Quantile
( )
( )
√
√ [ (
√
)] ( ) ( )
{ ( )}
√
Where,
𝒇( ) ( )| | | |
( ) & ( ) ( )
√ √
Therefore
(a) is a consistent and asymptotically normal estimator of
√
(b) The asymptotic variance of is
Hence option (a), (b), (c) and (d) are correct.
30. Consider a M/M/1 queue with arrival rate and service rate . Let =0 and denote the
queue length at time t. Which of the following statements are true?
(a) ( )admits a stationary distribution if and only if
(b) The stationary distribution of the process ( ) is geometric, when it exists
(c) ( )=1 for all k if
( )
(d) ( )= for all k if
Ans: b, c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page16

CSIR NET Dec-2019
Single Choice Questions [SCQ]
1. In factorial design, treatmennt combinations of there treatments A, B and C are allotted to
2 bocks of 4 plot each. Suppose the key block is as follows.
Key block: (1) a bc abc
Then the confounded treatment combination is
(a) AB
(b) AC
(c) BC
(d) ABC
Ans: c
2. Let be two state Markov chain with state space S and transition matrix

[ ]

Assuming the expected return time to 0 is

(a)
(b)
(c)
(d) 3
Ans: a
3. Let and be a random sample of size 2 from Uniform[0, ] distribution, Define
. What is the confidence coefficient of the confidence interval [ ]
for
(a) 0.6285
(b) 0.7535
(c) 0.8333
(d) 0.9500
Ans: b
4. Let ( ) be a random sample from a distribution with probability density
function
( )
{
̅ be the sample mean and ( ) . Then which of the following
statements is correct?
(a) ( ) is sufficient for
(b) ( ) is unbiased for
(c) ̅ is unbiased for
(d) ̅ is sufficient for
Ans: c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page17

5. Suppose data ( )( ) ( ) are generated as follows:
Bernoulli ( ) and | ( ) Define
( ) {
Then, which of the following is a correct linear regression model for ( ) ( | ) in
the sense that the true ( ) is obtained for some valued of the parameters for all
(a) ( )
(b) ( )
(c) ( ) ( )
(d) ( ) ( )
Ans: d
6. The draw a sample of size n ( ) using a without replacement scheme from a finite
population of size N, the first unit is chosen using PPS( )
scheme and the remaining ( ) units are drawn using SRSWOR. Then the probability that
is include in the sample is
(a)
(b)
(c)
(d)
Ans: d
7. Subjected to the conditions

the minimum value of the function is

(a) 10
(b) 0
(c) -25
(d) -15
Ans: d
8. There are three urns each with balls of two colours. Contains 2 white balls and
3 black balls, contains 3 white balls and 2 black balls and contains 5 white balls and 5
black balls. An urn is chosen at random and a ball is drawn from that urn at random. What is
the probability that that was chosen given that the ball picked is black in colour?
(a)
(b)
(c)
(d)
Ans: b
[ ] [ ]
9. Let X be a real- valued random variable such that [ ] and . Then
which of the following is correct?
[ ]
(a) [ ]
(b) [ ] ( [ ])
PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page18
 (c) ( )
(d) almost suerly
Ans: b
10. Let X and Y be independent Exponential random variables with means and respectively
with . Let ( ) denote the density function of . Then for
(a) ( ) ( ) ( )
(b) ( )
(c) ( ) ( )

(d) ( ) {

Ans: c
11. Let ( )( ) ( ) be a sample from a Bivariate Normal
( ) distribution with all parameters unknown. For testing against
If you use the usual t-test and your observed sample correlation coefficient is 0,
then what is p-value?
(a) 0
(b) 0.05
(c) 0.5
(d) 1
Ans: d
12. Suppose ( ) ( ) ( ) are i.i.d observation from the Uniform distribution on the unit
square [0, 1] [0, 1]. What is the probability that the rank correlation between the and
values is 1?
(a) 0
(b)
(c)
(d)
Ans: d
Multiple Choice Questions [MCQ]
13. Let be i.i.d Normal random variables with mean and variance where
is unknown. Then which of the following statements are correct?
(a) is unbiased for
(b) is unbiased for
(c) is unbiased for ( )
(d) ( ) is unbiased for ( )
Ans: a, c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page19

14. Let be i.i.d observation with ( ) and ( ) Then which of the
following statements are correct?
(a) ∑ ( ) is consistent for
(b) ∑ ( ̅) is consistent for
(c) ∑ is consistent for
(d) ∑ ( ̅) is consistent for
Ans: b, c, d
15. Let be i.i.d with common unknown continuous distribution function ( )
Where is the unique median of F. Define
{ and ∑
For testing against Which of the following statements are correct?
(a) ( )
(b) Test based on is distribution-free under
(c) Right-tailed test based on is unbiased
(d) The sequence of right-tailed tests based on is consistent
Ans: a, c, d
16. Let X be a discrete random with sample space X and probability mass function
( ) Consider testing the hypothesis
( ) against
( )
based on a single observation X. Then which of the following statements are correct?
(a) The test with critical region is most powerful of its size
(b) The test with critical region is unbiased at level
(c) If the p-value of the most powerful test is 0.6
(d) There exists a nonrandomized test of size 0.05
Ans: a
17. Suppose that ( ) and ( ) are independent where is positive
definite. Let be a fixed p-vector such that and define || || Then which
of the following random variables has a distribution, possibly after being scaled by a
constant factor?
(a)
(b) ( )
(c)
(d) ( )
Ans: b, c, d
18. Consider a Markov Chain with state space S. Let d(k) denote the period of state Which
of the following statements are correct ?
( ) ( )
(a) For if such that and and is recurrent, then is
recurrent
( ) ( )
(b) For if such that and then ( ) ()

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page20

 ( )
(c) For if such that then cannot transient
( )
(d) For if such that and is null recurrent then is positive
recurrent
Ans: a, b
19. Let be Markov Chain with state space such that the transition probability
are given by

for 0,1,2, …, where Then which of the following statements are correct?
(a) The Markov chain is irreducible
(b) The Markov chain aperiodic
( )
(c)
(d) The Markov chain is positive recurrent.
Ans: a, b, c, d
20. Consider a Balance Incomplete Block Design (v, b, r, k, ) of v treatments and b blocks of k
plots each. Let N be the incidence matrix of the design. Then which of the following
statements are correct?
(a) ( ) ( )
(b)
(c) ( )
(d) ( )
Ans: b, c, d
21. To estimate the population total Y =∑ , where are study variables and N is
the size of the finite population, a sample of xize n is dawn using PPSWR( )
scheme. if ̂ is the Hansen-Hurwitz estimator ofr Y then which of the following are
correct?
(a) ̂ is unbiased for Y
(b) Var ( ̂ ) = {∑ }

(c) Var ( ̂ )= ∑ { (̂ )}

(d) Var ( ̂ )= ∑ ∑ ( )
( )
Ans: a, b, c
22. Consider the random effect model where
N (0, ) and ~ i.i.d. N (0, ) are all independent of each other. The parameter
space for the model is ( ) ℝ [0, ) [0, ). Let ̂ and ̂ be the usual
unbiased estimators of and respectively. Then, which of the following evens can
happen with positive for some probability for some parameter values?
(a) ̂ is negative
(b) ̂ is negative
(c) ̂ is negative
(d) ̂ is negative

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page21

 Ans: b
23. Suppose in single service queue, customers arrive at a Poisson rate of one pre ten minutes,
and the service time is Exponential at a rate of one service per five minutes let be the
probability that there are n customers in the system in system in steady state. Then which of
the following statements are correct?
(a) for all n 0
(b) The expected number of customers in the system is 1 in steady state
(c) The expected number of customers in the system is 2 in steady state
(d) The expected amount of time a customers in the system is 1 in steady state is 10 minutes
Ans: a, b, d
24. Let {( ): n 1} and (X, Y) be random variables, on (Ω, ,P). Then which of the
following statements are correct?
(a) If X almost surely, Y almost surely, then X + Y in distribution
(b) If X in probability, Y almost surely, then X +Y in distribution
(c) If X in probability, Y in probability, then X +Y in distribution
(d) If X in distribution, Y in distribution then X +Y in distribution
Ans: a, b, c
25. Suppose the conditional p. d. f. of a random variable X given θ is

f (x|θ) = {
where the prior distribution of θ is Uniform (0, 1). Based on a single observation x from X,
which of the following statements are correct?
(a) The Bayes estimate for θ under squared error loss function is – x log
(b) The Bayes estimate for θ under squared error loss function is –
(c) The Bayes estimate for θ under absolute error loss function is x
(d) The Bayes estimate for θ under absolute error loss function is –
Ans: b, d
26. Let X and Y be real-valued independent random variable on Ω. Then which of the following
statements are correct?
(a) E[ ( )] = E[cos(tX)] E[cos(uY)] - E[sin(tX)] E[sin(uY)] for all t, u,
(b) If X~ N (2,1) and Y~ N (0,2), then Var(X + Y) = 3 where N ( ) represents Normal
distribution with mean and variable
(c) {X = } {Y = b} = ϕ for all a, b
(d) P({X = } {Y = b}) = P({X = }) P({Y = b}) for all a, b
Ans: a, c, d
27. Suppose , are i. i. d. Uniform (θ, 2θ), θ > 0. Let ( ( ) ) = min { } and
( ( ) ) = max { }. Then which of the following statements are correct?
(a) ( ( ) ( ( ) ) is jointly sufficient and complete for θ
(b) ( ( ) ( ( ) ) is jointly sufficient but not complete for θ
( )
(c) is a maximum likelihood estimator for θ
(d) ( ) is a maximum likelihood estimator for θ
Ans: b, c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page22

28. A random variable T has symmetric distribution if T and –T have the same distribution. Let
X and Y be independent random variables. Then which of the following statements are
correct?
(a) If X and Y have the same distribution then X-Y has a symmetric distribution
(b) If X ( ) and Y ( ), then 2X-3Y has symmetric distribution
(c) If X and Y have the same symmetric distribution, then X+Y has a symmetric distribution
(d) If X has a symmetric distribution, then XY has a symmetric distribution
Ans: a, b, c, d
29. Given data ( ) where and not all are identical, the sample
linear regression model is fit. Let be the diagonal element of the hat
matrix ( ) , where is the corresponding model matrix. Then which of the
following are possible for some choice of n and ?
(a)
(b)
(c)
(d)
Ans: c, d
30. Consider the following system with three independent components .
𝑢

𝑢 𝑢

Suppose that the failure probability of each components is , and let ( ) be the probability
that the whole system is still functioning. Then which of the following statements are
correct?
(a) ( )
(b) ( )
(c) ( )
(d) ( )
Ans: a, c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page23

CSIR NET June-2019
Single Choice Questions [SCQ]
1. Let , … be a sequence of independent normally distribute random variables with mean
1 and variance 1. Let N be a Poisson random variable with men 2, independent of ,…
be then, the variance of +…+ is
(a) 3
(b) 4
(c) 5
(d) 9
Ans: c
2. There are two sets of observations on a random vector (X, Y). Consider a simple linear
regression model with an intercept for regressing Y and X. let ̂ be the least squares estimate
of the regression coefficient obtained from the i-th (i = 1,2) set consisting of observations
( ). Let ̂ be the least squares estimate obtained from the pooled sample of size
. If it is known that ̂ ̂ which of the following statements is true?
(a) ̂ ̂ ̂
(b) ̂ may lie outside ( ̂ ̂ ),but it cannot exceed ̂ ̂
(c) ̂ may lie outside ( ̂ ̂ ),but it cannot be negative
(d) ̂ can be negative
Ans: d
3. Suppose and are sample multiple correlation coefficients of on and
on respectively. Which of the following is possible?
(a)
(b)
(c)
(d)
Ans: c
4. A simple of size n = 2 is drawn from a population on size N = 4 using probability population
to size without replacement sampling scheme, where the probabilities proportional to size are
1 2 3 4
0.4 0.2 0.2 0.2
The probability of inclusion of unit 1 in the sample is
(a) 0.4
(b) 0.6
(c) 0.7
(d) 0.75
Ans: c
5. Suppose that ( ) has a bicariate density
+ , where and are respectively, the densities of bivariate normal
distributions ( ,∑) and ( ,∑), with = ( ), = ( ) and ∑ the 2 2 identity
matrix. Then which of the following is correct?
(a) X and Y are positively correlated

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page24

 (b) X and Y are negative correlated
(c) X and Y are uncorrelated but they are not independent
(d) X and Y are independent
Ans: a
6. The maximum value of the objective function z = 5x + 2y under the linear constraints
x 0, y 0, x y and 2 x+ y 4 is
(a) 14
(b) 20
(c) 25
(d) 17
Ans: b
7. There are 30 questions in ascertain multiple choice examination paper. Each question has 4
options and exactly one is to be marked by the candidate. Three candidates. A, B, C. marks
each of the 30 questions at random independently. The probability that all the 30 answers of
the three students match each other perfectly is
(a)
(b)
(c)
(d)
Ans: d
8. Let be i.i.d. random variables having a continuous distribution function.
Then P( | = max ( )) equals
(a)
(b)
(c)
(d)
Ans: c

9. Consider a Markov Chain with state space {0,1,2,3,4} and transition matrix

( )
( )
Then equals
(a)
(b)
(c) 0
(d) 1
Ans: c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page25

10. Consider the function f(x) defined as ( ) , . For what value of c is f a
probability density function?
(a) ( )
(b) ( )
(c) ( )
(d) ( )
Ans: a
11. Suppose (X, Y) following bivariate normal distribution with means , standard deviations
and correlation coefficient p, where all the parameters are unknown. Then testing :
is equivalent to testing the independence of
(a) X and Y
(b) X and X – Y
(c) X + Y and Y
(d) X + Y and X
Ans: d
12. A random sample of size 7 is drawn from a distribution with p.d.f.

( ) { ( )

and the observations are 12, - 54, 26, - 2, 24, 17, - 39. What is the maximum likelihood
estimate of ?
(a) 12
(b) 24
(c) 26
(d) 27
Ans: d
Multiple Choice Questions [MCQ]
13. Consider a classification problem between two classes having densities
( ) 1; 0 x 1 and
(x) = 1+ ( ); 0 x 1
respectively. Assume that the prior probabilities of the two classes are equal. Which of the
following are true?
(a) The Bayes classifier classifier an observation to class-1 if x ( )
(b) A randomly chosen observation from class- 1 is misclassified with probability
(c) A randomly chosen observation from class- 2 is misclassification probability
(d) The average misclassification probability of the Bayes classifier is
Ans: a, b, c, d
14. Let … ~ i.i.d N (μ, ), where μ and are unknown. Consider a SRSWR

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page26

 { … }of size n from … , as well as a SRSWOR { … } of size n
from … . To estimate Y = ∑ ̅ define ̂ = ∑ and ̂
∑ . Then which of the following are true?
( )
(a) The conditional variance of ̂ … is ∑ ( ̅)
( )
( )
(b) The conditional variance of ̂ give … is ∑ ( ̅)
( )
( )
(c) The unconditional variance of ̂ is
(d) The unconditional variance of ̂ is
Ans: b, d
15. In a design with four treatments and three blocks of 2 plots each, treatments A and B are
allotted to blocks 1 and 3 and treatments C and D are allotted to block 2. Hence the resulting
design
(a) is incomplete
(b) is connected
(c) is non-orthogonal
(d) has all elementary treatment contrasts estimable
Ans: a, c
16. Consider a system with two components whose lifetimes are i.i.d. exponential with hazard
rate . Let and be the hazard functions of the system if the components are put in series
and parallel, respectively. Then which of the following are true?
(a) (t) (t) for all t > 0
(b) (t) < λ for all t > 0
(c) (t) < λ for all t > 0
(d) is a strictly increasing function of t
Ans: a, b, d
17. You want to invest Rs.1000 in companies I and II. If the market is good, company I will
declare dividend of 50% while company II will declare 30%. If the market is bad, company I
will declare dividend of 10% while company II will declare 20%. The prediction is that
market will be good with probability 0.4 and bad with probability 0.6. the investment that
maximizes expected dividend is
(a) Rs.1000 in company I and nil in company II
(b) Nil in company I and Rs.1000 in company II
(c) Rs.500 in each of the two companies
(d) Rs.600 in company I and Rs.400 in company II
Ans: a
18. Consider an M/M/1 Queue with arrival rate λ. For t > 0, let be the number of arrivals upto
(and including) t. for k 1, let the last arrival time of the k-th customer. Let =t-
be the time elapsed after the last arrival and = - t be the waiting time from t to next
arrival. Then which of the following are true?
(a) is an unbounded random variable
(b) ( ) = 1/ λ
(c) is an unbounded random variable

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page27

 (d) ( ) = 1/ λ
Ans: c, d
19. Suppose a normal Q - Q plot is drawn using a reasonably large sample from an
unknown probability distribution. For which of the following distribution would you expects
Q – Q plot to be convex (J – shaped)?
(a) Beta (5, 1)
(b) Exponential (1)
(c) Uniform (0, 1)
(d) Geometric (1/2)
Ans: b, d
20. ( ) and X are random variables on a probability space. Suppose that converges
to X in probability. Which of the following are true?
(a) E(| | ) 0
(b) ( ) P(X x) for all x
(c) E (min( | |)) 0
(d) | | 0 with probability
Ans: c
21. Consider a simple symmetric random walk on integers, where from every state i you move to
and with probability half each. Then which of the following are true?
(a) The random walk is aperiodic
(b) The random walk is irreducible
(c) The random walk is null recurrent
(d) The random walk is positive recurrent
Ans: b, c
22. Consider a Markov Chain with state space {0, 1, 2} and transition matrix

P=

( )
Then which of the following are true?
( )
(a) =0
( ) ( )
(b) = =0
( )
(c) =
( )
(d) =
Ans: b, d
23. Suppose X, Y are i.i.d. Binomial (n, p) random variables. Which of the following are true?
(a) X + Y ~ Bin (2n, p)
(b) (X,Y) ~ multinomial (2n; p, p)
(c) Var(X – Y) = E ( )
(d) Cov(X + Y, X – Y) =0
Ans: a, c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page28

24. Let … i.i.d N (0, ), where (> 0) is unknown. Consider a class of estimators
∑ for . Then which of the following are true?
(a) is consistent for if
(b) is unbiased for if
(c) is consistent for if
(d) MSE of is minimized for if
Ans: a, b, c, d
25. Let … be i.i.d random variable with pdf
( )
( ) {

For testing : = 2 against : = 3, the following tests are suggested.

Test 1: Reject if and only if max { } 1
Test 2: Reject if and only if max { } 1
Let denote the probability of type I error for test i, i = 1, 2. Which of the following are
true?
(a) Test 1 is a level 0.05 test
(b) =1
(c) Test 2 is unbiased
(d) Test 1 is unbiased
Ans: b, c
26. Let be the square region with four vertices ( ,0), (0, ), ( ,0) and ( 0, ), where > 0
is unknown. Suppose that ( ) follows the uniform distribution on Which of the
following statements are true?
(a) X and Y are uncorrelated
(b) X and Y are independent
(c) | | + | | cannot exceed
(d) If ( ) and ( ) are two observations
Ans: a, c, d
27. Let … (n > 5) be i.i.d. with p.d.f. , which is symmetric about having
bounded support. Let ( ) ( ) ( ) be the order statistic of the random
variables … which of the following statements are correct?
(a) ( ) – and ( ) have the same distribution
(b) ( ) – and ( ) have the same distribution
(c) The distribution of ( ) is symmetric about
(d) E[ ( ) ( ) ] is same for all k = 1, 2, …, n
Ans: b, c, d
28. X following a p-variete (p > 2) distribute with E (x) = μ and Var (x) = ∑. Which of the
following are true?
(a) E ( )=∑+
(b) E ( )= + trace( )
(c) (( )∑ ( )) = p

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page29

 (d) P[( )∑ ( ) ] for 0
Ans: a, b, c, d
29. Consider a logistic regression model with a single regress or , where the log odds ratio has
intercept and slop . Let be the ratio of the odds of success when to that when
. Which of the following are true?
(a) =
(b) =1
(c) =
(d) =
Ans: a, d
30. Let … ~ i.i.d. N(0, ) and ~i.i.d. N (0, ) for i = 1, 2, …, 10; j = 1, 2, …,n
independently of … .Define = . As an estimator of , the statistic
= ∑ ∑ is
(a) unbiased
(b) consistent as n
(c) unbiased only when = 0
(d) consistent as n only when =0
Ans: c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page30

CSIR NET Dec-2018
Single Choice Questions [SCQ]
1. If X ~ (0, I) and is an idempotent matrix with rand (A) = k < p, then which of the
following statements is correct?
(a)
(b)
(c) ( )
(d) ( )
Ans: d
2. A sample of size n( ) is drawn from a population of N( ) units using PPSWR sampling
scheme, where is the probability of selecting unite in a draw, 0 < <1 = , …, N,
and ∑ = 1. Then the inclusion probability
(a) 1 +( )
(b) 1 – ( )
(c) 1 ( ) ( ) ( )
(d) 1 ( ) ( ) ( )
Ans: d
3. In a experiment with two blocks and factors A, B, C and D, one block contains the
following treatment combinations a, b, c, ad, bd, cd, abc, abcd. Which of the following
effects is confounded?
(a) ABC
(b) ABD
(c) BCD
(d) ABCD
Ans: a
4. In an airport, domestic passengers and international passengers arrive independently
according to Poisson processes with rates 100 and 70 per hour, respectively,. If it is given
that the total number of passengers (domestic and international) arriving in that airport
between 9:00 AM an 11:00 AM on a particular day was 520, then what is the conditional
distribution of the number of domestic passengers arriving in this period?
(a) Poisson (200)
(b) Poisson (100)
(c) Binomial ( )
(d) Binomial ( )
Ans: c
5. Let X 0 be random variable on ( ) with (X) = 1. Let A be an event with 0 < p
(A) < 1. Which of the following defines another probability measure on ( )?
(a) Q(B) = P(A )
(b) Q(B) = P(A )
(c) Q(B) = E(A )

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page31

 ( ) ( )
(d) Q(B) = {
( )
Ans: c
6. Let X and Y be i.i.d. random variable uniformly distributed on (0, 4). Then P(X >Y X < 2Y)
is
(a)
(b)
(c)
(d)
Ans: a
7. Suppose { } is a Markov Chain with 3 states and transition probability matrix

( )

Then which of the following statements is true?

(a) } is irreducible
(b) } is recurrent
(c) } does not admit a stationary probability distribution
(d) } has an absorbing sate
Ans: d
8. Given the observations 0.8, 0.71, 0.9, 1.2, 1.68, 1.4, 0.88, 1.62 from the uniform distribution
on (θ – 0.2, θ + 0.8) with < θ < , which of the following is a maximum likelihood
estimate for θ?
(a) 0.7
(b) 0.9
(c) 1.1
(d) 1.3
Ans: b
9. To test the hypotheses against using the test statistic T, the proposed test procedure is
not to support if T is large. Based on a given sample, the p-value of the test statistic is
computed to be 0.05 assuming that the distribution of T is N(0,1) under . If the distribution
of T under is the t-distribution with 10 degrees of freedom instead, the p-value will be
(a) 0.05
(b) < 0.05
(c) 0.05
(d) > 0.05
Ans: d
10. Let ( ) ( ),…,( ) be n independent observations from a vicariate continuous
distribution. Let be the product moment correlation coefficient and be the rank
correlation coefficient computed based on these n observations. Which of the following
statements is correct?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page32

 (a) 0 implies 0
(b) 0 implies 0
(c) 1 implies 1
(d) 1 implies 1
Ans: c
11. Consider a linear model
= + + for i = 1, 2 and
= + + for i = 3, 4
Where are independent with E( ) = 0, Var ( )= > 0 for i = 1,…, 4. and ,…,
. Which of the following parametric functions is estimable?
(a) +
(b)
(c) +
(d) + +
Ans: c
12. Suppose X ~ Cauchy (0, 1). Then the distribution of is
(a) Uniform (0, 1)
(b) Normal (0, 1)
(c) Double exponential (0, 1)
(d) Cauchy (0, 1)
Ans: d
Multiple Choice Questions [MCQ]
13. Suppose n( ) units are drawn from a population of N (> n) units sequentially as follows. A
random sample of size N is drawn from U (0, 1). The k-th population unit is
selected if < , k = 1, 2, …, N, where = 0 and = number of units selected out
of first k – 1 units for each k = 2, 3, …, N then
(a) The probability on inclusion to the 2nd unit in the sample is
( )
(b) The probability of inclusion of the 1st and 2nd unit in sample is ( )
st nd
(c) The probability of not including the 1 unit and including the 2 unit in the sample is
( )
( )
(d) The probability of including the 1st unit but not including the 2nd unit in the sample is
( )
( )
Ans: a, b, c
14. Consider a block design with three blocks and four treatments A, B, C and D where A. B and
D where only A and B are allotted to block-1, only A, B and D are allotted to block-2 and
only C is allotted to block-3. Then the resulting block design is
(a) incomplete and ton connected
(b) incomplete and not balanced
(c) balanced and connected
(d) neither balanced nor connected
Ans: a, b, c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page33

15. Suppose X is a positive random variable with the following probability density function
( ) ( ) , for and Then the hazard function
of X for some choices of and can be
(a) an increasing function
(b) a decreasing function
(c) a constant function
(d) a non-monotonic function
Ans: a, b, c, d
16. A parallel system has n( 1) identical components. The lifetimes of the lifetime of the n
distributed exponential random variable with mean 1. If the lifetime of the system is denoted
by X, then which of the following statements are true?
(a) The mode of X is 0 for some n.
(b) The mode of X is less than or equal to n for all n.
(c) The mean of X is greater than equal to 1 for all n.
(d) The median of X is greater than 100 for some n.
Ans: a, b, c, d
17. Suppose ABC is a triangle on the xy-plane with centroid D. Which of the following points
can NEVER be a minimizer of the function 7x – 10y +1 as (x, y) runs over the triangle ABC?
(a) A
(b) B
(c) C
(d) D
Ans: d
18. Suppose is a random sample from the uniform distribution of (0, 2) and =
max { } for every positive integer n. Then which of the following statements are
true
(a) 2 almost surely
(b) 2 in probability
(c) 2 destitution
(d) converges in distribution to normal distribution
√
Ans: a, b, c
19. Let be i.i.d. N(0, 1) random variable. Let = + +…+ Which
of the following statements are correct?
(a) ~ N(0,1) for all n 1.
√
(b) For all > 0, P (| | ) 0 as n
(c) with probability 1.
(d) P( √ ) P (Y ) x ,
Ans: c, d
( )
20. Let {X} be a Markov chain with state space S. for any i, j S, let denote the n-step
transition of probability of going from i to j. Let d(i) denote the period of state i (i ).
Which of the following statements are correct?
( )
(a) If d (i) = d (j) then > 0.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page34

 ( ) ( )
(b) If d (i) = d (j) then > 0 and > 0 for some
( ) ( )
(c) If > 0 and > 0 for some n, m 1, then d (i) = d (j).
( )
(d) > 0 implies d (i) = d (j).
Ans: c
21. Consider a Markov chin with transition probability matrix P given by

( ).

( )
For any two states i and j, let denote the n-step transition probability of going from i to j.
Identify correct statements.
( )
(a) = 2/9.
( )
(b) = 0.
( )
(c) = 1/3.
( )
(d) = 1/3.
Ans: a, d
22. Suppose that ( ) follows a vicariate distribution with common marginal distribution F
and Corr ( ) = 0. Then which of the following statements are correct?
(a) F = Uniform (0, 1) are independent.
(b) F = Berrnoulli (θ) are independent.
(c) F = Discrete uniform {-1, 0, 1} are independent.
(d) F = N (0, 1) are independent
Ans: b
23. Let be a random sample from the distribution with p.d.f.

(x) = {

where > 0. Then which of the following are true?

(a) ∏ is sufficient for
(b) ∑ is sufficient for
(c) ∏ is a maximum likelihood estimate for .
(d) ∑ is maximum likelihood estimate for
Ans: a, c, d
24. X is a discrete random variable on {- 2, - 1, 1, 2} with probability mass functions [X = x]
given below
X -2 -1 2 2
0.05 0.6 0.3 0.05
0.2 0.4 0.2 0.2
The aim is to test : against : . Which of the following statements are
correct?
(a) The test procedure with critical region {x = 2} is a most powerful test of size 0.05
(b) The test procedure with critical region {x = - 2} is a most powerful test of size 0.05

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page35

 (c) The test procedure with critical region {x = - 1} is not a most powerful test of its size
(d) The test procedure with critical region {x = 1} is not a most powerful test of its size
Ans: a, b, c, d
25. Suppose is a random sample from uniorm distribution of ( ), where
is an unknown parameter. Let ( ) ( ) ( ) be the corresponding order-
statistics. Which of the following are 100 (1 α)% confidence intervals for ?
(a) (- ( ) )
(b) ( ( ) )
(c) ( )
(d) ( )
Ans: a, b, d
26. Let be independent random variable following a common continuous
distribution F, which is symmetric abut 0. For i = 1, 2, …, n define

={ and

= rank of | | in the set {| | … | | }. Which of the following statements are correct?

(a) are independent and identically distributed
(b) are independent and identically distributed
(c) S = ( ) and R = ( ) are distributed
(d) Distribution of T= ∑ dose not dependent on the functional form of F.
Ans: a, c, d
27. Suppose Y|θ ~ Poisson (θ), θ > 0 and prior density of θ is given by (θ) ,
where > 0 and are hyper- parameters. Which of the following are true?
(a) Marginal distribution on Y is hyper geometric
(b) Posterior distribution of θ given Y = y is Gamma
(c) is a conjugate prior
(d) Bayes’ estimator of θ for squared error loss function is
Ans: b, c, d
28. Consider a linear model = + , where Disp ( ) = for some >
0. One needs to choose the design matrix X such that its elements take values in the set -1, 0,
1}. Now, consider the following three choices of X

[ ], [ ] and = [ ]

Which of the following statements are true?

(a) For all three choice of X, = ( ) is estimable
(b) For all three choices of X , ̂ ̂
and , the least squared estimates of and , are
uncorrelated for all i j
(c) is a better choice than
(d) is a better choice ttheen
Ans: a, b, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page36

29. Suppose that are i.i.d. N((0, 1). Which of the following statements are correct?
(a) P{ }=
(b) P{ }=
(c) P{ ( ) ( ) ( ) ( )} =
(d) P{ ( ) ( )} =
Ans: a, b, c, d
30. Consider a classification problem between two uniform distributions U(0, 1) and U((1,5).
Let (0 1) be the prior probability of the calls having U(0, 2) distribution. If we
consider the loss function , which of the following statements are correct?
(a) For < 1/3, the Bayes’ risk (i.e the average misclassification probability of the Bayes’
classifier) is smellier then 1/6
(b) For < 1/3 the Bayes’ risk is smaller than 1/6
(c) < 1/3 < 1/3 the Bayes’ risk is 1/6
(d) For all choices off , the Bayes’ classifier is unique
Ans: a, c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page37

CSIR NET June-2018
Single Choice Questions [SCQ]
1. Consider a Markov chain having state space with transition probability
matrix ( ) given by

[ ] then

( ) ( )
(a) ∑
( ) ( )
(b) ∑
( ) ( )
(c) ∑
( ) ( )
(d) ∑
Ans: b
2. Let b i.i.d. standard normal variables, which of the following is true?
√ | |
(a)
√

(b)
√ | |
( )
(c) ( )

(d)
Ans: b
3. Suppose that the lifetime of an electric bulb follows an exponential distribution with mean
hours. In order to estimate bulbs are switched on at the same time. After t hours,
( ) bulbs are found to be in functioning state. If the lifetimes of the other ( )
bulbs are noted as respectively, then the minimum likelihood estimate of is
given by
(a) ̂
∑
(b) ̂
∑ ( )
(c) ̂
∑ ( )
(d) ̂
Ans: d
4. be i.i.d. uniform ( ) variables, where are unknown parameters.
Which of the following is an ancillary statistic?
( )
(a)
( )
( ) ( )
(b)
( )
( )
(c)
( ) ( )

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page38

 ( ) ( )
(d)
( ) ( )
Ans: d
5. Consider the problem of estimation of a parameter on the basis of X, where
( ) . Under squared error loss, X has uniformly smaller risk
than that of kX, for
(a)
(b)
(c)
(d) No value of k
Ans: c
6. To test the equality of effects of 10 schools against all alternative form each school and note
their marks in a common examination. “Between sum of square” an “total sum of square”
are found to be 180 and 500 respectively. What is the p-value for the standard F-test?
(a) [ ]
(b) [ ]
(c) [ ]
(d) [ ]
Ans: d
7. The covariance matrix of a four dimension random vector X is of the form

[ ] where

If v is the variance of the first principal component, then

(a) v cannot exceed 5/4
(b) v cannot exceed 5/4, but cannot exceed 4/3
(c) v can exceed 4/3, but cannot exceed 3/2
(d) v can exceed 3/2
Ans: b
8. A simple random sample of size n will be drawn from a class of 125 students, and the mean
mathematics score of the sample will be computed. If the standard error of the sample mean
for “with replacement sampling” is twice as much as the standard error of the sample mean
for “without replacement” sampling , the value of n is
(a) 32
(b) 63
(c) 79
(d) 94
Ans: d
9. In a Latin Square Design the “error degrees of freedom” is 30. The “treatment degrees of
freedom” for any treatment is
(a) 4
(b) 5
(c) 6
(d) 7

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page39

 Ans: c
10. Suppose that | | | | Then the maximum value of is
(a) 1
(b) 2
(c) 3
(d) 4
Ans: c
11. A standard fair die is rolled until some face other than 5 or 6 turns up. Let X denote the face
value of the last roll, and
.
Then,
(a) ( )
(b) ( )
(c) ( )
(d) ( )
Ans: c
12. Let X and Y be i.i.d. uniform (0, 1) random variables. Let ( ) and
( )
Then (( ) ) is
(a) ½
(b) ¾
(c) ¼
(d) 2/3
Ans: c
Multiple Choice Questions [MCQ]
13. Consider a single server M/M/1 queue with arrival rate and service rate further assume
that Then, which of the following statements are true?
(a) Queue length becomes 0 infinitely many time intervals with probability 1
(b) Queue length becomes 0 in at most finitely many time intervals with probability 1
(c) Steady state exists for the queue
(d) ( ) where is the number of customers in the system at time t
Ans: a, c, d
14. Let be 20 observations in the interval [0,1]. Let ̅ and ̃ be the mean and the
median of these observation, and let ∑( ̅)
(a) If 15 observations are smaller than 0.3, then ̅ cannot exceed 0.5
(b) will be maximum if 10 of these observation are 1 and the rest are 0
(c) If all observation except one are smaller than 0.5, then ̅ cannot be smaller than ̃
(d) ̅( ̅)
Ans: a, b, d
15. A statistician has drawn a simple random sample of size 2 with replacement from 4 boys
with distinct heights. Let ̅ be the sample mean of their heights. Then another statistician
has drawn a simple random sample of size 2 without replacement from those 4 boys. Let ̅
be the sample mean of their height. Which of the following statements are correct?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page40

 (a) ( ̅ ̅ ) has larger variance than that of ( ̅ ̅ )
(b) ( ̅ ̅ ) has larger variance than that of ( ̅ ̅ )
(c) ( ̅ ̅ ) has larger smaller variance than that of ( ̅ ̅ )
(d) ( ̅ ̅ ) has larger smaller variance than that of ( ̅ ̅ )
Ans: a, b
16. In a data set with mean 2.5 and standard deviation 0.5.
(a) The median must be bigger than 2.5
(b) The median must be smaller than 2.5
(c) The median must be smaller than 3
(d) The median must be bigger than 2
Ans: c, d
17. Let F, h and m be the lifetime distribution function, the hazard function and the mean
resident on [ ] Assume that F is absolutely continuous. Which of the following
statements are true?
(a) ∫ ( )
∫ ( ( )
(b) ( ) ( )
(c) ( )is strictly increasing in t if the lifetime distribution is exponential with mean
(d) ( ) ( ) lifetime distribution is exponential with mean
Ans: b, d
18. Consider a Markov chain on state space with transition probability matrix P
given by

( )
Let ( ) be a stationary distribution of the Markov chain and d(1) denote the
period of state 1. Which of the following statements are correct?
(a) ( )
(b) ( )
(c)
(d)
Ans: a, d
19. Let be a sequence of i.i.d. random variables with ( ) ( ) Which
of the following are true?
(a) ∑ in probability
(b) ∑ in probability
(c) ∑ in probability
(d) ∑ in probability
Ans: b, d
20. Let X and Y be i.i.d exponential random variables with parameter 1. Define
( ). Which of the following are true?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page41

 (a) ( )
(b) is uniform on (0,1)
(c) are independent
(d) are uncorrelated, but dependent
Ans: a, b, c
21. Let be i.i.d. with probability mass function
( ) ( )
( ).
Which of the following statements are true?
(a) is sufficient statistic
(b) is a sufficient statistic
(c) is a sufficient statistic
(d) is a sufficient statistic
Ans: a, c, d
22. Suppose that for are i.i.d. ( ) are i.i.d.
( ). Assume further that the and the are independent. Let r be the
correlation coefficient computed from the bivariate data ( )( ) ( ) Then
( )
(a) has distribution (F-distribution with 1 and ) for all
√( )
(b) has distribution (t-distribution with 1 and ) for all
√
(c) has the distribution of the square of a Cauchy variable for
(d) has a beta distribution for all
Ans: a, b, c, d
23. Suppose X and Y are two independent exponential random with means and
respectively, where is unknown. Which of the following statements are true?
(a) is sufficient for
(b) Right-tailed test based on is UMP for testing against
(c) Left-tailed test based on is UMP for testing against
(d) UMP test does not exist for testing against
Ans: c, d
24. Consider a two- sample location problem with 6 and 8 observations from the first and
second population, respectively. Suppose that the distribution of the population is
( ) ( ) where F is a continuous distribution function with the
median at 0. Define T as the sum of the ranks of second sample in the combined sample For
the problem of testing against , which of the following statements
are true when all observations are independent?
(a) T is distribution free under
(b) It is appropriate to reject when T is small
(c) Observed value of T cannot be 85
(d) E(T)=60 under
Ans: b, c, d
25. We observe ( ) where is known but is an
known parameter, Note that when n=0, X is degenerate at 0. Suppose that n has a prior

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page42

 distribution which is Poisson with a known mean Which of the following statements
are correct?
(a) The posterior distribution of n is also Poisson but with a mean different from
(b) If X=0, the posterior distribution of n is Poisson with mean ( )
(c) The Bayes estimate of n has bias ( ) when
(d) The Bayes estimate of n has larger variance than the variance of the unbiased estimate

Ans: b, c
26. In the linear model :

Where ( ) and

[ ]

Let ( ̂ ̂ ̂ ) be the least squares estimate of ( ) Let

(a) ( ̂ ̂ ̂ ) is unique
(b) ∑ ̂ is the best linear unbiased estimate (BLUE) of ∑
(c) ∑ ̂ is the uniformly minimum variance unbiased estimate (UMVUE) of
∑
(d) ∑ ̂ is BLUE but not (UMVUE) of ∑
Ans: a, b, c
27. Let be i.i.d. ( ) variable where and both are unknown parameters.
Consider a confidence interval for , which is of the from
∑( ̅) ∑( ̅)
[ ] where Let be the cumulative distribution function of a
chi-square random variable with n degrees of freedom.Which of the following statements
are true ?
(a) It is possible of the find a 95% confidence interval of the from where ab=1
(b) If ( ) ( ) then it is the shortest 95% confidence interval.
(c) If it is the shortest 95% confidence interval, then a and b must satisfy the condition
( )
(d) If ( ) ( ) , then the expected length of a 95% confidence
interval of the form is ( )( )
Ans: a, d
28. Consider a two-class classification problem, where the densities of the two competing
classes are given by
( ) {
And
( ) {

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page43

 Let be the prior probabilities of these two classes. Now consider a classifier
which classifies an observation to class 1 if and to class 2 if .
(a) If then is the Bayes classifier
(b) If then is the Bayes classifier
(c) If then is the Bayes classifier
(d) If then the average probability of misclassification for is 3/8.
Ans: a, d
29. Let X and Y be two random variable with joint probability density function
( ) {
Which of the following statements are correct?
(a) X and Y are independent
(b) ( )
(c) ( )
(d) ( )
Ans: b, c, d
30. Let X and Y be two random variables satisfying
( ) ( ) ( ) ( )
Which of the following statements are correct?
(a) ( )
(b) ( )
(c) ( )
(d) ( )
Ans: c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page44

CSIR NET Dec-2017
Single Choice Questions [SCQ]
1. There are five empty boxes. Balls are placed independently one after another in randomly
selected boxes. The probability that the fourth ball is the first to be placed in an occupied
box equals
(a) ( )

(b) ( )

(c) ( )
(d) ( ) ( )
Ans: a
| |
| |
2. Let Ψ (t) = and ( ) = . Which of the following is true?
(a) Ψ is a characteristic function but is not
(b) is a characteristic function but Ψ is not
(c) Both Ψ and are characteristic function
(d) Neither Ψ and nor characteristic function
Ans: c
3. Consider a Markov chain { | } with state space {1, 2, 3} and transition matrix

. Than P ( | = 1) equal

( )
(a) 0
(b)
(c)
(d)
Ans: c
4. X, Y are independent exponential random variables with means 4 and 5, respectively. Which
of the following statements is true?
(a) X + Y is exponential with mean 9
(b) XY is exponential with mean 20
(c) Max (X,Y) is exponential
(d) Min (X, Y) is exponential
Ans: d
5. Let and be a random sample of size two from a distribution with probability density
| |
function ( ) + (1 θ) , , where θ { }. If the
√
observed values of and are 0 and 2 , respectively, then the maximum likelihood
estimate of θ is

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page45

 (a) 0
(b)
(c) 1
(d) Not equal
Ans: a
6. Let X be a random sample of size 1 from a Cauchy distribution with probability density
function ( ) ( ) , where ( ). For testing test :θ=
( )
1 against : θ = 0, the following test is suggested. Reject if √ > C, otherwise do
not reject . What is the value of C so that the power the power of the test is 0.5?
(a)
(b) 0
(c) ( )

(d) A solution of √
Ans: b
7. Let be a random sample from uniform (0, 3θ), θ > 0. Define =
max }. Which of the following is NOT true?
(a) is consistent for θ
(b) is unbiased for θ
(c) is a sufficient statistic
(d) is complete
Ans: b
8. Consider the following regression problem = ; i = 1,…, n here , i = 1, 2, …, n,
are i.i.d N (0, 1) random variable. It is assumed that α 0 and β is known. If ̂ is the MLE
of α, which of the following statements is true?
(a) (̂ ) α
(b) (̂ )
(c) (̂ )
(d) ( ̂ )
Ans: d
9. Suppose (0, ∑ ), (0, ∑ ), where and are independently distributed. If
and ∑ ∑ are positive definite then which of the following statements is necessarily
true?
(a) ∑ ∑
(b) ∑ ∑
(c) ∑ ∑
∑
(d) ∑
~
Ans: b
10. Suppose we draw a random sample of size n from a population of size N, where 1 < n < N,
using simple random sampling without replacement scheme. Let P be the population

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page46

 proportion of units possessing particular attribute and p be the corresponding sample
proportion. Which of the following is unbiased estimator for P (1 – P)?
(a) p(1 – p)
(b) p(1 – p)
( )
(c) p(1 – p)
( )
( )
(d) p(1 – p)
( )
Ans: c
11. Let X and Y be independent exponential random variables. If E[X] =1 and [ ] then P
(X > 2Y|X > Y) is
(a)
(b)
(c)
(d)
Ans: d
12. A parallel system consists of n identical components. The lifetimes of the components are
independent identically distributed uniform random variables with mean 30 hours and range
60 hours. If the expected lifetime of the system is 50 hours, then the value of n is
(a) 3
(b) 4
(c) 5
(d) 6
Ans: c
Multiple Choice Questions [MCQ]
13. Suppose A,B,C are events in a common probability space with P (A) = 0.2 P(B)= 0.2
P (C) = 0.2 ,P (A ) = 0.1, P(A )= 0.1, P (B ) = 0.1. Which of the following are
possible values of P (A )?
(a) 0.5
(b) 0.3
(c) 0.4
(d) 0.9
Ans: a, c
14. Let S be the set of all 3 3 matrices having 3 entries equal to 1 and 6 entries equal to 0. A
natrux M is picked uniformly at random from the set S. then
(a) P {M is nonsingular} =
(b) P {M has rank 1} =
(c) P {M is identity} =
(d) P {trace (M) =0} =
Ans: a, b
15. Consider a Markov chain with five states {1, 2, 3, 4, 5} and transition matrix

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page47

 P=

( )
Which of following are true?
(a) 3 and 1 are in the same communicating class
(b) 1 and 4 are in the same communicating class
(c) 3 and 2 are in the same communicating class
(d) 2 and 5 are in the same communicating class
Ans: b, d
16. Which of the following are correct?
(a) If X and Y are N(0,1) then is N (0,1)
√
(b) If X and Y are independent N(0,1) then has t-distribution
(c) If X and Y are independent Uniform N(0,1) then is Uniform (0,1)
(d) If X is Binomial (n, p) then n – X is Binomial (n,1 – p)
Ans: b, d
17. For n 1, let be a Poisson random variable with mean . Which of the following are
equal to
√
∫
(a) ( )
(b) ( )
(c) ( )
(d) ( )
Ans: a, c
18. Let be a random sample from ( ), probability density function or a
probability mass faction. Define ∑ ( ̅ ) , where ̅ = ∑ . Then is
unbiased for θ if
(a) (x) and θ > 0

(b) (x) , ,θ>0

√ √
(c) (x) θ>0
(d) (x) θ>0
Ans: a, b
19. Let { } be a sequence of independent random variables where the distribute of is normal
∑
with mean and variance n for n =1, 2, …Define ̅ and
∑ ∑ Which of following are true?
̅
(a) E ( ) = E ( ) for sufficiently large n

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page48

 (b) Var ( ) < Var ( ̅ ) for sufficiently large n
(c) ̅ is consistent for μ
(d) ̅ is sufficient for μ
Ans: a, b
20. Consider a Cauchy population with probability density function ( ) ,
( )
Let be a random sample from the above
population. Which of the following confidence intervals for θ have confidence coefficient 1 –
α (0 < α < 1)?
( ) ( )
(a) [ ]
( ) ( )
(b) [ ]
( ) ( )
(c) [ ]
( ) ( )
(d) [ ]
Ans: a, b, c, d
21. Let be a random sample from an unknown continuous distribution faction F
with median θ. Let count the number of i for which > 0. Consider the problem of
testing : θ = 0 against : θ = -1 based on the test statistic . Which of the following are
true?
(a) The distribution of is independent of F under
(b) Left-tailed test based on in consistent against
(c) Left-tailed test based on in consistent against
(d) Left-tailed test based on has the p-value P[ ] under
Ans: b, c
22. Consider the following regression problem = + + here , … ,
are i.i.d N (0, ) random variables. If ̂ and ̂ are the least square estimators of
, respectively, then which of the following statements are correct?
(a) E( ̂ ) =
(b) E( ̂ ) =
(c) Var( ̂ ) > Var( ̂ )
(d) Cov ( ̂ ̂ ) < 0
Ans: a, b, c
23. Let be a random sample from N (θ, 1), θ [ - 100, 100] and be define by
={
Suppose ̂ and ̃ denote the MLEs of θ based on { } and on { },
respectively. Which of the following statements are true?
(a) (̂ ) = 0
(b) (̃ ) = 0
(c) ̂ is a consistent estimator of θ
(d) ̃ is a consistent estimator of θ
Ans: a, b, c, d
24. Suppose we have a random sample of size n (n ) from the density

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page49

 (x) = {
If the prior of λ is an exponential distribution with mean 1, then which of the following
statements are correct?
(a) The posterior distribution of λ is an exponential distribution
(b) The Bayes estimator of λ w.r.t. the squared error loss function exists and is unique
(c) The Bayes estimator of λ w.r.t the absolute error loss faction exists and is untrue
(d) The Bayes estimator of does not exist
Ans: b, c
25. Suppose are i.i.d random vectors from (0, ∑). Let , E (∑ )=
c and E (∑ ) = A. Which of the following statements are necessarily true?
(a) c =
(b) (∑ ) follows a chi-squared distribution
(c) (∑ ) and (∑ ) are independently distributed for 1 n–1
(d) ∑
Ans: c
26. Suppose there are k groups each consisting if N boys. We want to estimate the mean age μ of
these kN boys. Fix 1 < n < N and consider the following two sampling schemes.
1. Draw a simple random sample without replacement of size k n out of all k N boys.
2. From each of the k groups draw a simple random sample with replacement of size n. let
̅ and ̅ be the respective sample mean ages for the fwo schemes. Which of the
following are true?
(a) E ( ̅ ) = μ
(b) E ( ̅ ) = μ
(c) Var ( ̅ ) may be less than Var ( ̅ ) in some cases
(d) Var ( ̅ ) = Var ( ̅ ) if all the group means are same
Ans: a, b, c
27. Consider a BIBD (v, b, r, k, λ) with k = 5. Let N =(( )) be the incidence matrix, where =
number of times ith treatment appears in jth block, 1 i let C = rI
which of the following are true?
(a) C has a characteristic root 0
(b) Rank of N is v
(c) The above BIBD is connected
(d) Trace of the C is 4b
Ans: a, b, c, d
28. Twenty identical items are put in a life testing experiment stating at time 0. The failure times
of the items are recorded in a sequential manner. The experiment stops if all the items fail or
a pre-fixed time T > 0 is reached, whichever is earlier. If the lifetimes of the items are
independent identically distributed exponential random variables with mean θ. where 0 < θ
10, then which of the following statements are correct?
(a) The MLE of θ always exists
(b) The MLE of θ may not exist
(c) The MLE of θ is an unbiased estimator of θ, if it exists
(d) The MLE of θ is bounded with probability 1, if it exists

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page50

 Ans: a, d
29. Arrival of customers in a shop is a Poisson process with intensity λ = 2. Let X be the number
of customers entering during the time interval (1, 2). Which of the following are true?
(a) P (X = 0|X + Y = 12) = ( )
(b) X and Y are independent
(c) X + Y is Poisson with parameter 6
(d) X – Y is Poisson with parameter 8
Ans: a, b
30. Consider an M/M/1 queue with interracial time having exponential distribution with mean
and service time having exponential distribution with mean . Which of the following are
true?
(a) If 0 < λ < μ then the queue length has limiting distribution Poisson (λ - μ)
(b) If 0 < μ < λ then the queue length has limiting distribution Poisson (μ - λ)
(c) If 0 < λ < μ then the queue length has limiting distribution which is geometric
(d) If 0 < μ < λ then the queue length has limiting distribution which is geometric
Ans: c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page51

CSIR NET June-2017
Single Choice Questions [SCQ]
1. Let X be a random sample from an exponential distribution with mean 1/λ. If λ has a prior
distribute with probability density function g(λ) = {
Then the Bayes estimator of 1/λ with respect to the squared error loss function is
(a)
(b)
(c) X
(d)
Ans: d
2. Consider the linear statistical model = + + ; i = 1, 2, … , ; j = 1, 2, … , n where
μ is unknown ae independently and identically distributed as N (0, ), are
independently and identically distributed as N (0, ); and ; are independent for all i
and j. note that is the treatment effect. Suppose , are total
sum of squares, total treatment sum of squares and error sum of squares, respectively. To
test: which of the following statements is not true?
(a) The sum of squares identity is
(b) ( )

(c) Under , ~ ( )
( )
(d) E ( ) = n (a – 1) ( )
Ans: b or c or d*
3. Suppose ( ) follows a bivariate normal distribution with E ( ) = E ( ) = 0, V ( ) =
V ( ) = 2 and Cov( ) = 1. If ϕ (x) = ∫ dy, then P [ > 6] is equal
√
to
(a) Φ( - 1)
(b) Φ( - 3)
(c) Φ(√ )
(d) ( √ )
Ans: d
4. Consider the problem of drawing a sample of size 2 from a finite population of size 20. The
sampling is done with replacement using probability proportional to size sampling scheme.
The normal size measures , …, are given by = , i = 1 ,…, 20, = , i = 11 , …,
20. The expected number of distinct units drawn is
(a)
(b)
(c)
(d)

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page52

 Ans: d
5. A box contains 40 numbered red balls and 60 numbered black balls. From the box, balls are
drawn one by one at random without replacement till all the balls are drawn. The probability
that the last ball drawn is black equals
(a) 1/100
(b) 1/60
(c) 3/5
(d) 2/3
Ans: c
6. , … are independent identically distributed random variable having common density f.
assume f (x) = f( - x) for all x . Which of the following statements is correct?
(a) ( ) 0 in probability
(b) ( ) 0 almost surely
(c) P( ( ) )
√
(d) ∑ has the same distribution as ∑ ( )
Ans: d
7. Let denote the number of accidents up to time t. assume that { } is a Poisson process
with intensity 2. Given that there are exactly 5 accidents during the time period [20, 30],
what is the conditional probability that there is exactly one accident during the time period
[15, 25]?
(a)
(b) 20
(c)
(d)
Ans: a
8. X and Y are independent random variables each having the density
f (t) ,
Then the density function of for is given by
(a)
(b)
(c)
(d)
Ans: a
9. If we interchange two columns of a Latin square design (LSD). Then the new design is
(a) an LSD
(b) a completely randomized design (CRD) but not an LSD
(c) a randomized block design (RBD) but not an LSD
(d) a balanced incomplete block design (BIBD) but not an LSD
Ans: a

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page53

10. Consider the LPP:
Minimize x subject to Ax = b, x 0. Where
A=[ ], b = [ ],
C=( ) and x = ( ) . Using the revised simplex method with
current basis as [ ], which of the following statements is correct?
(a) The next entering variable is
(b) The solution corresponding to the current basis is optimal
(c) The next entering variable is
(d) The next entering variable is
Ans: c
11. Suppose { }, n 2, is a random sample from the distribution with probability

density function f (x; θ) { ( )

with θ > 0. Then the method of moment’s estimator of θ

(a) does not exist
(b) is ∑ ( )
(c) is ∑ ̅)
(
(d) is ∑ ( )
Ans: c
12. Let for n 5 be a random sample from the distribution with probability density
( )
faction f (x; θ) {
for θ > 0. The confidence coefficient of the confidence interval
[ ] For θ, is
(a) 0.5
(b) 0.75
(c) 0.95
(d) 1 -
Ans: b
Multiple Choice Questions [SCQ]
13. Let } be a random sample from the probability density function
| |
f (x; θ) = ; where θ . Which of the following statements are
correct?
(a) The maximum likelihood estimator of θ is ∑
(b) ∑ is a sufficient statistic for θ
(c) The maximum likelihood estimator of θ is a function of a sufficient statistics
(d) There does not exist a uniformly most powerful test for the following testing problem: :
θ = 0 vs : θ 0

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page54

 Ans: c, d
14. Consider the problem of testing : θ = 1 vs :θ where θ is the mean of a Poisson
random variable. Let X and Y are a random sample from Poisson (θ) distribution. Consider
the following test procedure:
Reject if either X = 0 or (X = 1 and X + Y ); otherwise accept . Which of the
following test procedure:
(a) P[type I error] = +
(b) P[type I error] =
(c) Size of the test is +
(d) Power of the test is +
Ans: a, d
15. Suppose } is a random sample from the distribution with probability density
function f(x). Consider the following testing problem using likelihood ratio test (LRT).
| |
: f (x) = vs : f (x) =
√
Which of the following statements are correct?
(a) There does not exists any LRT.
(b) The rejection is a function of | |, … , | |
(c) The rejection region is a function of ,…, .
(d) The rejection region is of the from {∑ (| | ) }.
Ans: b, c, d
16. Let c be a constant. Let X,Y be random variables with joint probability density function
f (x; y) = {
Which of the following statements are correct?
(a) c =
(b) c = 8
(c) X and Y are independent
(d) P(X =Y) = 0
Ans: b, d
17. Let { } be i.i.d uniform ( -1 , 2) random variables. Which of the following
statements are true?
(a) ∑ almost surely
(b) { ∑ ∑ } almost surely
(c) = 2 almost surely
(d) = -1 almost surely
Ans: b, c, d
18. Let { } be a Markov chain on {0, 1, 2, …} with
, , , , i 1, = 0 otherwise. Which of the following
statements are correct?
(a) { } is recurrent

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page55

 (b) { } is transient
(c) P( )>0
(d) P( )>0
Ans: b, d
19. Which of the following statements are correct?
(a) For a finite state Markov chain there is at least one transient state.
(b) For a finite state Markov chain there is at least one stationary distribution.
(c) For a countable sate Markov chain, every state can be transient.
(d) For an aperiodic countable state Markov chain there is least one stationary distribution.
Ans: b, c
20. Suppose X follows an exponential distribution wit parameter λ > 0. Fix a > 0. Define the
random variable Y by Y = k, if ka X < (k + 1)a, k = 0, 1, 2, …
Which of the following statement s are correct?
(a) P (4 < Y < 5) = 0
(b) Y follows an exponential distribution
(c) Y follows a geometric distribution
(d) Y follows a Poisson distribution
Ans: a, c
21. Consider a finite population of size N. Let be the sample mean based on a sample of size n
under simple random sampling with replacement (SRSWR) scheme. Let be the sample
mean based on a stratified random sample of size n where the samples are drawn from each
of 4 strata using SRSWR scheme under proportional allocation. Then which of the following
are sufficient conditions for Var( ) = Var( ) to hold?
(a) Strata sizes are same
(b) Strata total are same
(c) Strata means are same
(d) Strata variances are same
Ans: c
22. Consider a balanced incomplete block design (BIBD) with parameters b, v, r, k, λ where each
of the b blocks contains k treatments out of a set of v treatments, each treatment occurs r
times in the design and each pair of treatments occurs λ times. A new design is formed by
replacing all the treatments in each block by its complementary set. Then which of the
following are true for the new design?
(a) It is a BIBD
(b) Each treatment occurs (b – r) times
(c) Each pair of treatments appears in the same block (b – r + λ) number of times
(d) bk = vr
23. Suppose the random variable X has the following probability density function
( )
( )
f (x) {
Where a > 0, < < . Which of the following statements are correct? The hazard
faction of X is
(a) an increasing function for all a > 0
(b) an decreasing function for all a > 0
(c) an increasing function for some a > 0

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page56

 (d) an decreasing function for some a > 0
Ans: c, d
24. Consider the problem:
Maximize 2 + 3
Subject to
+ 1, + 1, + 1,
1 and 0 for - = 1, 2, 3, 4.
For the optimum value is
(a) equal to 8
(b) between 8 and 9
(c) greater than or equal to 7
(d) less than or equal to 7
Ans: c, d
25. Le ( ) be an optimal solution to the problem of minimizing

Subject to the constraints

300
500
400
200
0, 0
Which of the following not possible vague for any ?
(a) 300
(b) 400
(c) 500
(d) 600
Ans: b, c, d
26. Let be i.i.d random variables with common continuous distribution function F(x
) and let be i.i.d random variable with common continuous distribution
function F (y ). Consider the problem of testing
: = vs : > .
Let be the ranks of , respectively in
the combined sample. Define
∑ ∑ Which of the following statements are true?
(a) E( ) = E( ) under
(b) E( ) = 52.5 under
(c) cannot be 27
(d) If we use right-tailed test based on , then the observed value is significant at
5% level of significance.
Ans: a, b, c, d
27. In a football league, the goals scored by home teams over 380 matches have the following
frequency distribution.
Number 0 1 2 3 4 5
of goals
Frequency 92 121 91 50 19 7

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page57

 The average goals scored by home teams are 1.49. We want to test
: Goal distribution is Poisson. Based on observations the value of the -statistic for
goodness of fit is 1.27. Given
= 1.64, = 1.15, = 12.59 and = 11.07. Which of the following
are true?
(a) : is not rejected at 5% level of significance.
(b) -statistic has 5 degrees of freedom
(c) Under , the maximum likelihood estimate (MLE) of the rate parameter of Poisson
1.49
(d) Under , in a game, the MLE of the probability that home team will score at most one
goal is 2.49
Ans: a, b, c, d
28. Consider the model Y = Xβ + ,

Where Y = | |, X = (( )) and β = | |, =| |.

E ( ) = 0 and D =( ) = , p < n. Let ̂ be the solution of X β = Y. Which of the

following are true?
(a) If is estimable then ̂ is the best linear unbiased estimator (BLUE) of β
(b) All linear parametric functions are estimable if and only if rank (X) > p
(c) If Rank (X) < p then some linear parametric functions are not estimable
(d) ( ̂) ( ̂ ) (n – p) is an unbiased estimator of
Ans: a, c, d
29. Let and be independent random variable each having N(μ, ) distribution, where
μ , > 0. Let 0 θ < 2 and A =( ). Put =( ) =A with
( ) . Which of the following statements are correct?
(a) in distribution if and only if μ = 0.
(b) in distribution if and only if μ = 0, θ = 0.
(c) are Gaussian.
(d) may be correlated.
Ans: a, c
30. Let be two i.i.d (0, ∑) random variables with rank (∑) = p. suppose A is a p p
symmetric matrix of rank r, and = A. which of the following statements are correct?
(a) A ~
(b) A + ~2
(c) A + ~2
(d) A + ~
Ans: a, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page58

CSIR NET Dec-2016
Single Choice Questions [SCQ]
1. There are two boxes. Box 1 contains 2 red balls and 4 green balls. Box 2 contains 4 red balls
and 2 green balls. A box is selected at random and a ball is chose randomly form the selected
box. If the ball turns out to be red, what is the probability that Box 1 had been selected?
(a)
(b)
(c)
(d)
Ans: b
2. For any two events A and B, which of the following relations always holds?
(a) (A ) + (A )+ ( )
(b) (A ) + (A )+ ( )
(c) (A ) + (A )+ ( ) 1
(d) (A ) + (A )+ ( )
Ans: a
3. Suppose customers arrive in a shop according to a Passion press with rate 4 per hour. The
shop opens at 10:00 am. If it is given that the second customer arrives at 10:40 am, what is
the probability that no customer arrived before 10:30 am?
(a)
(b)
(c)
(d)
Ans: a
4. Let is a random sample from a distribution with probability density function
f (x) = 3 ( ) (x), where
( )
( )( ) {
What is the probability density function g (y) of Y = min { }?
(a) g(y) = ( ) (y).
(b) g(y) = 1 – ( ) ( ) (y).
(c) g(y) = ( ) ( ) (y).
(d) g(y) = ( ) ( ) (y).
Ans: d
5. are independent and identically distributed N (θ,1) random variable, where θ
takes only integer values i.e θ {…,-2,-1,0, 1, 2, …}. Which of the following is the maximum
likelihood estimator of?
(a) ̅
(b) Integer closest to ̅

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page59

 (c) Integer part of ̅ (Largest integer ̅)
(d) Median of ( ).
Ans: b
6. For a random variable X, with E(X) > 0, the coefficient of variable p is define as ( )
where is the variable of X, suppose are independent samples from a normal
population with mean 2 and unknown coefficient of variation . It is desired to test : 5
against : 5. The likelihood ratio test is of the from: Reject if
(a) ∑( ) > C.
(b) ∑( ) < C.
∑( ̅)
(c) ̅
> C.
∑( ̅)
(d) ̅
< C.
Ans: a
7. ( ), ( ), … , ( ) are data on X- cultivable land in a district and Y- the area
actually under cultivation, both measured in square feet. Let ̂, ̂ be the least squares
estimators of α, β in the model Y = α + βx + , where is the random error. If the data are
converted to square meters, then
(a) ̂ may change but ̂ will not.
(b) ̂ may change but ̂ will not.
(c) Both ̂ and ̂ may change.
(d) Neither ̂ nor ̂ will change.
Ans: a
8. Suppose in a one-way analysis of variance model, the sum of squares of all the group means
is 0 (Assume that all the observations are not same). Then the value of the usual F-test
statistic for testing the equality of means
(a) Cannot be determined from the above information.
(b) is undefined.
(c) is 0
(d) is 1.
Ans: c
9. Let ( ) be a random vector with mean
∑ Let of the first principal component
∑ is
(a) The vector of all the eigenvalues of ∑.
(b) The eigenvector corresponding to the smallest eigenvalue of ∑
(c) The eigenvector corresponding to the largest eigenvalue of ∑.
(d) The vector of all the eigenvalues of ∑
Ans: c
10. (v, b, r, k, λ) are the standard parameters of a balanced incomplete block design (BIBD).
Which of the following (v, b, r, k, λ) can be parameters of a BIBD?
(a) (v, b, r, k, λ) = (44, 33, 9, 12, 3)
(b) (v, b, r, k, λ) = (17, 45, 8, 3 1)
(c) (v, b, r, k, λ) = (35, 35, 17, 17, 9)

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page60

 (d) (v, b, r, k, λ) = (16, 24, 9, 6, 3)
Ans: d
11. Consider an M/M/1 Queue with arrival rate λ and service arte μ with μ > λ. What is the
probability that no customer exited the system before time 5?
(a)
(b)
(c) ( )

(d) ( )
Ans: a
12. A simple random sample (without replacement) of size n is drawn from a finite population of
size N ( ). What is the probability that the 4th population unit is included in the sample but
th
the 6 population unit is not included in the sample?
( )
(a) ( )
( )
(b) ( )
( )( )
(c) ( )
(d)
Ans: b
Multiple Choice Questions [MCQ]
13. A fair die is thrown two times independently. Let X, Y be the outcomes of these two throws
and Z = X + Y. Let U be the remainder obtained when Z is divided by 6. Then which of the
following statement(s) is/are true?
(a) X and Z are independent
(b) X and U are independent
(c) Z and U are independent
(d) Y and Z are not independent
Ans: b, d
14. A and B play a game of tossing a fair coin. A starts the game by tossing the coin once and B
the tosses the coin twice, followed by A tossing the coin once and B tossing the coin twice
and theism continues until a head turns up. Whoever gets the first head wins the game. Then
(a) P(B wins) > P (A wins )
(b) P(B wins) = 2P (A wins )
(c) P(A wins) > P (B wins )
(d) P(A wins) = 1 - P (A wins )
15. Consider the Markov chin with state space S = {1, 2, …, n} where n > 10. Suppose that the
transition probability matrix P = (( )) satisfies
> 0 if | | is even
= 0 if | | is odd
Then
(a) The Markov chain is irreducible.
(b) There exists a state i which is transient.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page61

 (c) There exists a state i with period d (i) = 1.
(d) There are infinitely many stationary distributions.
Ans: b, d
16. Let { ; i 1} be a sequence of independent random variables each having a normal
distribution with mean 2 and variance 5. Then which of the following are true
(a) ∑ converges in probability tot 2.
(b) ∑ converges in probability tot 9.
(c) ( ∑ ) converges in probability tot 4.

(d) ∑ ( ) converges in probability tot 0.

Ans: a, b, c, d
17. Let X be a random variable with a certain non-degenerate distribution. Then identify the
correct statements
(a) If X has an exponential distribution then median (X) < E (X)
(b) If X has a uniform distribution on an interval [a, b], then E (X) < median (X).
(c) If X has a Binomial distribution then V(X) < E (X)
(d) If X has normal distribution, then E (X) < V (X)
Ans: a, b
18. Suppose the probability mass function of a random variable X under the parameter
θ = and θ = ( ) are given by
X 0 1 2 3
p ( ) 0.01 0.04 0.5 0.45
p ( ) 0.02 0.08 0.4 0.5
Define a test ϕ such that
Φ (x) = 1 if x = 0, 1
= 1 if x = 2, 3
For testing θ: = against θ: = . The test ϕ is
(a) A most powerful test at level 0.05
(b) A likelihood ratio test at level 0.05
(c) An unbiased test
(d) Test of size 0.05
Ans: a, b, c, d
19. are independent and identically distribution as N (μ, ). <μ< , > 0.
Then
( ̅)
(a) ∑ is the Minimum Variance Unbiased Estimate of
( ̅)
(b) √∑ is the Minimum Variance Unbiased Estimate of σ
( ̅)
(c) ∑ is the Minimum likelihood Estimate of
( ̅)
(d) √∑ is the Minimum likelihood Estimate of σ
Ans: a, c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page62

20. Let be independent and identically distribution random variable each following
uniform (1 – θ, 1+ θ) distribution, θ > 0. define
( ) = min { },
( ) = max { } and ̅ = ∑ .Which of the following is true?
(a) ( ( ) ̅ ( ) ) is sufficient true θ
(b) ( ( ) - ( )) is unbiased for θ
(c) ∑ ( ) is unbiased for
(d) ∑ ( ̅ ) is unbiased for
Ans: a, c
21. Suppose are independent and identically distributed with common continues
distribution function F(x) and are independent and identically distributed with
common contiguous distribution function F(x – θ). Also suppose and are independent
for i, j. consider the problem of testing : θ > 0. Let = Rank ( ), α = 1, 2, … , m and
= Rank ( ), β = 1, 2, … ,n among , . Define U =
∑ ( ), where Ψ (a, b) = 1 if a < b,
= 0 if a b.
Which of the following are true?
(a) P[ ]=( )
under
(b) U and ∑ are linearly related
(c) E (U) under
(d) Right tailed test based on U is appropriate for testing against
Ans: a, b, c, d
22. θ is the probability of obtaining a head in the toss of a coin. The coin is tossed three times
and we record
Y= 1 if all the three tosses result in heads
Y= 2 if all the three tosses result in tails
Y= 3 otherwise
If the prior density of θ is Beta (α, β), and ̂ is the posterior mean of θ given Y = i, for i =
1, 2, then
(a) ̂ > ̂
(b) ̂ < ̂
(c) The posterior density of θ given Y = 3 is a Beta density
(d) The posterior density of θ given Y = 3 is not a Beta density
Ans: a, d
23. Suppose are independent and identically distributed standard nomal random
variables, and ( ) . If A is an idempotent k k matrix, then which of the
following statements are true?
(a) A and (I – A) are independent.
(b) A and (I – A) are identically distributed if k is even and trace (A) = .
(c) A follow a gamma distribution of A 0.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page63

 (d) (I – A) follow a chi-squared distribution if A 1.
Ans: a, b, c, d
24. For a data set ( , ), ( , ), …, ( , ) the following two models wre fitted using least
square method.
Model 1
= ; i = 1, 2, …n
Model 2
= ; i = 1, 2, …n
Let ̂ , ̂ be least square estimates of form model 1 and be the least square
estimates from model 2.
Let A=∑ ( (̂ ̂ ))
B=∑ ( ( ))
Then
(a) A B
(b) A B
(c) It can happen that A = 0 but B > 0
(d) It can happen that B = 0 but A > 0
Ans: a, d
25. Let ϕ (x, y ; ) be the density of biracial normal distribution with mean vector ( ) and

variance-Covariance matrix ( ). Consider a random vector ( ) having density

( ( ) ( )). Then
(a) Marginal distribution of both X and Y is standard normal.
(b) Covariance (X, Y) = 0
(c) X and Y are independent.
(d) (X, Y) has a bivariate normal distribution.
Ans: a, b
26. Suppose ̅ is the sample mean of the study variables corresponding to a sample of size n
using simple random sampling with replacement scheme and ̅ is the sample mean of the
study variables corresponding to a sample of size n using stratified random sampling with
replacement scheme under proportional allocation. Which of the following is/are sufficient
condition/conditions for Var( ̅ ) = Var ( ̅ )?
(a) All the stratum sizes are equal
(b) All the stratum total are equal
(c) All the stratum means are equal
(d) All the stratum variances are equal
Ans: c
27. We are given some balanced incomplete block designs (BIBDs) with parameters (v, b, r, k, λ)
such that λ = 1 and k = 1 (are fixed). With which of the following values of v can one
construct such a BIBD?
(a) v = 15
(b) v = 23

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page64

 (c) v = 25
(d) v = 28
Ans: a, c
28. A date ste gave a 95% confidence interval (2.5, 3.6), for the mean μ of a normal population
with known valance. Let < 2.5 be a fixed number. If we use the same data to test : μ =
vs :μ
(a) Would be necessarily rejected at α = 1.
(b) Would be necessarily rejected at α = .025
(c) For α = .1, the information is not enough to draw a conclusion
(d) For α = .025, the information is not enough to draw a conclusion
Ans: a, d
29. Suppose T follows exponential distribution with unit mean. Which of the following
statement(s) are correct?
(a) The hazard function of T is a constant function.
(b) The hazard function of is a constant function.
(c) The hazard function of is the identify function.
(d) The hazard function of √ is the identity function.
Ans: a, d
30. Consider the linear programming problem (LPP) maximize z = 3x + 5y
Subject to x + 5y 1
2x + 2y 5
x x, y 0.
Then
(a) The LPP does not admit any feasible solutions.
(b) There exists a unique optimal solution to the LPP.
(c) There exists a unique optimal solution to the dual problem.
(d) The dual problem has an unbounded solution
Ans: b, c

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page65

CSIR NET June-2016
Single Choice Questions [SCQ]
1. Let X and Y be independent and identically distributed random variables such that P (X = 0)
= P (X = 1) = . Let Z = X + Y and W = | |. Then which statement is not correct?
(a) X and W are independent.
(b) Y and W are independent.
(c) Z and W are uncorrelated.
(d) Z and W are independent.
Ans: c or d
2. Let { } and { } be two independent pure birth processes with birth rate
respectively. Let = + . Then
(a) { } is not a pure birth process.
(b) { } is not a pure birth process with birth rate .
(c) { } is not a pure birth process birth rate min ( )
(d) { } is not a pure birth process birth rate
Ans: b
3. Let ~ N (0, 1) and let
={
Then identify the correct statement.
(a) corr ( ) = 1.
(b) Does not have N (0, 1) distribution.
(c) ( ) has a bivariate normal distribution.
(d) ( ) dose not have a bivarite normal distribution.
Ans: d
4. Hundred (100) tickets are marked 1, 2, …, 100 and are picked from tithes tickets and are
given to for persons A,B,C and D . What is the profanity that a gets the ticket with the
largest value (among A, B, C, D) and D gets the ticket with the smallest value ((among
A, B, C, D)?
(a)
(b) 1/6
(c) ½
(d) 1/12
Ans: d
5. Consider arrive at an ice cream parlour account to a Poisson process with rate 2. Service
time distribution has density function f(x) = {
Upon being served a customer may rejoin the queue with probability 0.4, Independently of
new arrivals; also a returning customer’s service time is the same as that of a new arriving
customer. Customers behave independently of each other. Let X (t) = number of customers in
the queue at time t. which among the following is correct?
(a) {X (t)} grows without bound with probability 1.
(b) {X (t)} has stationary distribution given by = ( ) ( ) , k = 0, 1, 2, …

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page66

 (c) {X (t)} has stationary distribution given by = (0.1) ( ) , k = 0, 1, 2, …
(d) {X (t)} has stationary distribution given by = (0.4) ( ) , k = 0, 1, 2, …
Ans: a

6. Let ~ ( ∑) where = (1, 1, 1) and ∑ =( ). The value of c such that and

+ are independent is
(a) –2
(b) 0
(c) 2
(d) 1
Ans: c
7. Let be a random sample from N (θ, 1), where θ {1,2}. Then which or the
following statement about the maximum likelihood estimator (MLE) of θ is correct?
(a) MLE of θ does not exist
(b) MLE of θ ̅
(c) MLE of θ exist but it is not ̅
(d) MLE of θ in an unbiased estimator of θ
Ans: c
8. Let be denote a random sample from a N( ) distribution. Let be known
and (>0) be unknown. Let be an upper ( ) percentile point of a
distribution. The a 100 (1 – α)% confidence interval for is given by
(∑ ) (∑ )
(a) (( )
) ( )
∑ ( ) ∑ ( )
(b) (( )
) ( ) ( ) ( )
∑ ( ̅) ∑ ( ̅)
(c) ( ( )
) ( )
∑ ( ) ∑ ( )
(d) ( ( )
) ( )
Ans: d
9. In the context of testing of statistical hypotheses, which one of the following statements is
true?
(a) When testing a simple hypothesis against an alternative simple hypothesis , the
likelihood ratio principle leads to the most powerful test.
(b) When testing a simple hypothesis against an alternative simple hypothesis ,
P [rejecting is true] + P [rejecting is true] = 1.
(c) For testing a simple hypothesis against an alternative simple hypothesis ,
randomized test is used to achieve the desired level of the power of the test.
(d) UMP tests for testing a simple hypothesis against an alternative composite always
exist
Ans: a
10. Let be uncorrelated observations with common variance and expectations given
by ( ) = , ( ) = and ( ) = + , where , are unknon parameters. The best
linear unbiased estimator of + is

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page67

 (a)
(b)
(c) ( ).
(d) ( ).
Ans: c
11. A sample of size n ( ) is drawn without replacement from a finite population of size N,
using an arbitrary sampling scheme. Let denote the inclusion probability of the i-th unit
and , the joint inclusion probability of units i and j, 1 i j N. Which of the following
statements is always true?
(a) ∑
(b) ∑ i N
(c) > 0 for all i, j, 1 i j N
(d) > 0 for all i, j, 1 i j N
Ans: a
12. Consider a series system with two independent components. Let the component lifespan have
exponential with density ( ) {
If n observation on lifespan of the component are available and ̅ ∑ ,
Then the maximum likelihood estimator of the reliability of the system is given by
(a) ( ̅ )
(b) 1 - ( ̅ )
(c) ̅

(d) ̅

Ans: c
Multiple Choice Questions [MCQ]
13. Let (Ω, , P) be a probability space and A be an even with P(A) > 0. In which of the
following cases does Q define a probability measure on (Ω, F)?
(a) Q(D) = P (A ) D
(b) Q(D) = P (A ) D
( ) ( )
(c) Q(D) = {
( )
(d) Q(D) = P (D A) D
14. Suppose X and Y are independent an identically distributed random variable and let Z = X +
Y. then the distribution of Z is in the same family as that of X and Y if X is
(a) Normal
(b) Exponential.
(c) Uniform.
(d) Binomial.
Ans: a, d
15. Let be a random sample from the following bribability density function

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page68

 ( )
( )
f(x;μ,α) = { ( )

Here < < and α < 0. Then which of the following statements are correct?
(a) The method of moment estimators of neither α nor μ exist.
(b) The method of moment estimators of α exist and it is a consistent estimator of α
(c) The method of moment estimators of μ exist and it is a consistent estimator of μ
(d) The method of moment estimators of both α and μ exist, but they are not consistent
Ans: b, c
16. Let X be a random sample from a Poisson destruction with parameter λ. The parameter λ has
a prior distribution f (z); where f(z) = {
Under the squared error loss function, which of the following statements are correct?
(a) The Bayes’ estimator of is
(b) The posterior mean of λ is
(c) The posterior distribution of λ is gamma.
(d) The Bayes’ estimator of is ( )
Ans: a, b, c
17. Let be uncorrelated observations such that ( ) = + + = ( ),
( )= = ( ) and Var ( )= for i = 1, 2, 3, 4. Then, which of the following
statements are true?
(a) + + is estimable if and only if .
(b) An unbiased estimator of is [( ) ( ) ] .
(c) The best linear unbiased estimator of is ( ).
(d) The variable of the best linear unbiased of the best linear unbiased estimator of +
+ is .
Ans: a, b, c
18. Consider a linear regression model X + , where X is an n p matrix, rank (X ) = p,
( ) = 𝔻( ) = I, (.) stands for expectation, 𝔻 (.) denotes the variance covariance
matrix and I is the n-th order identity matrix. Define the n . matrix H = (( ))= X
( ) Then, which of the following are correct?
(a) 0 1, 1 n.
(b) If = 0 or 1 for some i, then = 0 for all j i
(c) The variance-covariance matrix of the vector of the predicted values ̂ (of Y) is H.
(d) For 1 n, if is the residual corresponding to i.e., – ̂ ̂ being the
predicted value of , that the valance of equals (1 ). (Here, is the i-th
component of ).
Ans: a, b, c, d
19. Let (X, Y) follow a vicariate normal distribution with mean vector (0, 0), and dispersion
Matrix ∑ = [ ], 0. Suppose Z = √ . Then which of the following statements are
correct?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page69

 (a) √ √
has a student-t distribution.

(b) √ √
has a student-t distribution.
(c) Z is symmetric about 0.
(d) E(Z) exists and equals zero.
Ans: a, c
( )
20. The hint probability density function of (X,Y) is f (x, y) = {
Which among the following are correct?
(a) X and Y are not independent
( )
(b) ( ) {
(c) X and Y are independent
( )
(d) ( ) {

Ans: a, b
21. Let be the result of the n-th roll of a fair die, n 1. Let =∑ and be the last
digit of for n 1 and = 0. Then, which of the following statements are correct?
(a) { } is and irreducible Markov chain
(b) { } is an periodic Markov chain
(c) P( ) as n .
(d) P( ) as n .
Ans: a, b, d
22. { } is a sequence of independent and identically distributed random variable with common
density function f(x) = {
{ } is a sequence of independent identically distributed random variable with common
density function g(y) = {
also { },{ } are independent families. Let = .Which anong the
following are correct?
(a) P( >0) >0.
(b) ∑ with probability 1
(c) ∑ with probability 1
(d) P( < 0) > 0.
Ans: a, c, d
( )
23. Suppose X is a random variable with following pdf f (x) = {
and 0 p 1. Then the hazard function of X is a
(a) constant function for p = 0 and p = 1
(b) constant function for all 0 p 1
(c) decreasing function for 0 < p < 1

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page70

 (d) non-monotone function for all 0 < p < 1
Ans: a, c
24. Let be a random sample from f (x; λ) = {
Here λ > 0 is an unknown parameter. It is desired to test the following hypothesis at level α >
0. We want to test : 1 vs : 1. Then which of the following ae true?
(a) UMP test is of the from ∑ < , with < for all n.
(b) UMP test is of the from ∑ < , with < for all n.
(c) UMP test is of the from ∑ < , with < for all n.
(d) UMP test is of the from ∑ < , with < for all n
Ans: b
25. Let be i.i.d. N (μ, 1). It is proposed to test : = 0 versus : = 0. Let (μ, α)
denote the power of the UMP test at μ of size α based of sample size n. Then which of the
following statements are correct?
(a) ( ) = 1 μ > 0, α > 0.
(b) ( ) = α n 1, α > 0.
(c) ( ) = 0 n 0, μ > 0.
(d) ( ) = 0 n 1, μ > 0.
Ans: a, b, c
26. A sample of size two is drawn from a population of 4 units using probability proportional to
size, sampling with replacement. The selection probabilities are = 0.2, = 0.3, = 0.1
and = 0.4 for units 1, 2, 3, and 4 in the population, respectively. Let the value of ,i = 1,
2, 3, 4. Let denote the inclusion probability of the i-th unit and the join inclusion
probability of units i and j, i < j, i, j = 1, 2, 3, 4. Then, which of the following statements are
correct?
(a) T = ( ) ∑ is an unbiased estimator of the population total, where the sum is over the
units in the sample
(b) = 0.36, .
(c) .
(d)
Ans: a, b, c
27. Maximize 3x + 4y subject to
x 0, y 0, x 3,
x + y 4, x + y + 5,
Which among the following are correct?
(a) The optimal value is 19
(b) The optimal value is 18
(c) (3, 2) is an extreme point of the feasible region.
(d) ( ) is an extreme point of the feasible region.
Ans: b, d
28. Let be a random sample from a uniform distribution on the interval
(θ – 1, ). Let = ̅ , the sample mean, = ̃ , the sample mean, and =

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page71

 be three estimators of θ. Then which of the following statements are correct?
(a) is consistent for θ.
(b) Both and are more efficient than
(c) All the three estimators are unbiased for θ
(d) is a sufficient statistic for θ.
Ans: a, c
29. Consider a balanced incomplete block design d with v treatments, b blocks. Replication r,
block size k and pairwise concurrence parameter λ. Assume the standard fixed effects model
for the data obtained through d. which of the following statements is(are) true?
(a) The design is connected if k 2.
(b) The inequality b v holds for d.
(c) The variance of the best linear unbiased estimator (BLUE) of a normalized treatment
contrast is a constant.
(d) The covariance between the BLUEs of two orthogonal treatment contrasts is zero.
30. Three types of components are used in electrical circuits 1, 2, 3 as shown below

A B

(Circuit 1)
B
A C

(Circuit 2)
C

(Circuit 3)
Suppose that each of the three of the three components fail with probability p and
independently of each other. Let = Prob (Circuit i does not fail); i = 1, 2, 3. For 0 < p < 1,
we have
(a)
(b)
(c)
(d)
Ans: a

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page72

CSIR NET Dec-2015
Single Choice Questions [SCQ]
1. From the six latter A, B, C, D, E and F. three letters are chosen at random with replacement.
What is the probability that either the word BAD or the word CAD can be formed from the
chosen letters?
(a)
(b)
(c)
(d)
Ans: d
2. Let X be a random variable which is symmetric about 0. Let F be the cumulative distribution
function of X. Which of the following statements is always true?
(a) F(x) + F (- x) = 1 for all x .
(b) F(x) - F (- x) = 0 for all x .
(c) F(x) + F (- x) = 1+ P(X = x) for all x .
(d) F(x) + F (- x) = 1 - P(X = x) for all x .
Ans: c
3. A set of N observations resulted in k distinct values , , … , with respective
frequencies , … , , so that ∑ = N. Another k observations , ,…, ones each,
so that the modified (new) sample pf size N + k has observation with frequency + 1.
(a) The new mean is necessarily less than or equal to the original mean.
(b) The new median is necessarily more than or equal to the original median.
(c) The new variable is necessarily less than or equal to the original variance.
(d) The new mode will be same as the original mode.
Ans: d
4. Let , and be uncorrelated observations with common unknown variance and
expectations given by
E ( ) = + + = E ( ),
E( )= =E( )
Where , and are unknown parameters. Define = ( ) and = =
√ √
( ). An unbiased estimator of is
(a) ( - ).
(b) ( + ).
(c) ( + ).
(d) ( + ).
Ans: b
5. Let , and be independent and identically distributed random variables with
commen distribution normal with mean μ and variance 2. If the prior distribution of μ is
normal with mean 0 and variance , then which of the following is true?
(a) The prior distribution is not a conjugate prior.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page73

 ∑
(b) Posterior mode of μ given , and .
∑
(c) Posterior median of μ given , and is .
∑
(d) Posterior variance of μ given , and is.( )
Ans: b
6. Let the n 1 vector follow an n-vitiate normal distribution with mean vector μ ( ) and
variance-matrix V ( , the order identity matrix). Also let A be a symmetric matrix of
order n. which of the following statements is ture?
(a) A always has a central chi-square distribution if and only if ( ) = AV
(b) A always has a central chi-square distribution if and only if =A
(c) The mean of A is Aμ + tr (AV), where tr (.) denotes the trace of a square matrix.
(d) A always has a central chi-square distribution with n degrees of freedom.
Ans: c
7. Consider the following Linear Programming Problem Max + subject to
5 +3 15
+ 1
2 +5 10.

The problem
(a) has no feasible solution.
(b) has infinitely many optimal solutions.
(c) has a unique optimal solution.
(d) has an unbounded solution.
Ans: c
8. Let ’s be independent random variables such than ’s are symmetric about 0 and Var ( )
= 2i – 1, for i . Then, ( )
(a) does not exist
(b) equals ½
(c) equals 1
(d) equals 0
Ans: d
9. Consider a randomize block design involving 3 treatments and 3 replicates and let denote
the effect of the treatment (i = 1, 2, 3). If denotes the variance of an observation,
wihich of the following statements is true?
(a) The variance of the best linear unbiased estimators (BLUE) of ( ) √ and (
) √ are equal.
(b) The covariance between the BLUE of and BLUE of is 2 /3
(c) The variance of the BLUE of (i ) 2 /3.
(d) The variance of the BLUE of ( ) is 2 /6
Ans: a
10. Let be two independent random variables taking values – 1 and + 1 with probability
each. Define = , = = ,…, = for n 3. Than

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page74

 (a) P ( = -1) =
(b) P ( = 1) =
(c) P ( = -1) =
(d) P ( = 1) =
Ans: b
11. Let , be a random sample from uniform (θ, 5θ) θ > 0. Define ( ) = min { ,
} and ( ) = max { , }. Maximum likelihood estimator of θ is
( )
(a)
(b) ( )
(c) ( )
( )
(d)
Ans: d
12. Consider the problem of testing : X ~ Normal with mean 0 and variance against : X ~
Cauchy (0,1). Then for testing against , the most powerful size α test
(a) dose not exist
(b) reject if and only |x| > where is such that the test is of size α
(c) reject if and only |x| < where is such that the test is of size α
(d) reject if and only |x|< or x > , where and are such that the test is
of size α
Ans: b
Multiple Choice Questions [MCQ]
13. Let ( ) be a Markov chain on the state pace S= {1, 2, …,23} with transition probability
given by
= = 2 i 22
= =
= = .
Then, which of the following statements are true?
(a) ( ) has a unique stationary distribution.
(b) ( ) is irreducible.
(c) ℙ ( )
(d) ( ) is recurrent.
Ans: a, b, c, d
14. A fair coin is tossed repeatedly. Let X be the number of Tails before the first Head occurs, let
Y denote the number of Tails observed between the occurrence of the first and the second
Heads. Let X + Y = N. Then, which of the following statements are true?
(a) X and Y are independent random variable with
( )
P (X = k) = P (Y = k) = {
(b) N has a probability mass function given by

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page75

 ( )
P{N = k} = {
(c) Given N = n, the conditional distribution of X and Y are independent
(d) Given N = n.
P{X = k} = {
Ans: a, d
15. Suppose that (X, Y) has a joint distribution with the marginal distribution of X being N (0, 1)
and E(Y X = x ) = for all x .. Then , which of the following statements are true?
(a) Corr (X,Y) = 0
(b) Corr (X,Y) > 0
(c) Corr (X,Y) < 0
(d) X and Y are independent.
Ans: b
16. An urn has 3 red and 6 black balls. Balls are drawn at random one by one without
replacement, the probability that second red ball appears at the fifth draw is
(a)
(b)
(c) 4 ( )
(d)
Ans: c
17. Let , denote a random sample from a distribution having a probability density
function f(x; θ) = θ , 0 < x < 1 , zero elsewhere; θ > 0. The set {( , ):
∑ ( ) } where c is a suitably chosen real number, is a uniformly most powerful
region for testing against when.
(a) : θ = 1 against : θ > 1.
(b) : θ = 1 against : θ 4.
(c) : θ = 1 against : θ 1.
(d) : θ = 1 against : θ 1.
18. Let , be independent and identically distributed, each having a uniform distribution
on (0, 1). Let = ∑ for n 1. Then, which of the following statements are true?
(a) 0 as n with probability 1.

(b) {{ } } =1.
(c) 0 as n with probability 1.

(d) {{ } } = 1.
Ans: a, d
19. Suppose ( ) is a random vector such that the marginal distribution of X and the marginal
distribution of Y are the same and each is normally distribute with mean 0 and variance 1.
Then, which of the following conditions imply independence of X and Y?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page76

 (a) Cov (X, Y) = 0
(b) aX + bY is normally distributed with mean 0 and variance for all real a and b.
(c) P (X 0, Y 0) = ¼.
(d) E[ ] = E[ ] E[ ] for all real s and t.
Ans: b, d
20. A experiment involving factors , , and each at two levels , coded 0 and 1 is
conducted in blocks of size 4 each. The block contents are as below:
Block I Block II

Block III Block IV

Then which of the following statements are true?

(a) The confounded effects are ,
(b) The confounded effects are ,
(c) The design is connected.
(d) The design is disconnected.
Ans: b, d
21. Let , are independently and identically distributed random variable , which
follow Bin(1, p). to test : p = vs : p = , with size α =0.01, consider the test
∑
Φ(t) = {
Then, which of the following statements are true?
(a) As n power of the test converges to
(b) As n power of the test converges
(c) As n power of the test converges
(d) As n power of the test converges 1
Ans: d
22. A system consists of 3 components averaged as in the figure below:

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page77

 Each of the components , has independent and identically distributed lifetimes
whose distribution is exponential with mean 1. Then, the survival function, S(t), of the
system is given by
(a) S(t) = for t > 0.
(b) S(t) = ( ) for t > 0.
(c) S(t) = ( ) for t > 0.
(d) S(t) =( ( ) ) for t > 0.
23. Let , be independent and identically distributed random variable with N(μ,1)
distribution. Assume that μ (0, ). Let ̂ be the MLE of μ. Then, which of the following
statements are true?
(a) ̂ = max ( ̅ , 0).
(b) ̂ is unbiased for μ
(c) ̅ is sufficient for μ
(d) ̂ is a consistent estimator of μ
Ans: a, c, d
24. Let , be a random sample from U(θ, θ + 1). If ( ) < ( ) <…< ( ) denote the
ordered values of , , then which f the following statements are true?
(a) ( ( ) , ( ) ) is a jointly sufficient statistic for θ
(b) ( ) is a sufficient statistic for θ
(c) ( ( ) ( ) ) is a jointly sufficient statistic for θ
(d) ( ) is a sufficient statistic for θ
Ans: a, c
25. A finite population has N units, labeled , and the value of a study variable on
unit is (i = 1, 2 …, N). Let Y = ∑ and ̅ = ∑ A sample of size n > 1 is
drawn from the population with probability proportional to size with replacement, with
selection probabilities , ; 0 < 1<, i= 1, 2,…,N and ∑ . = 1. Define T=
∑ , where the sum extends over the units in the sample. Then, which of the
following statements are true?
(a) T is an unbiased estimator of ̅ .
(b) T is an unbiased estimator of Y
(c) The variance of T is zero if is proportional to for all i, i = 1, 2,…, N.
(d) An unbiased estimator of the variance of T ∑ ( )
( )
Ans: b, c, d
26. Let , be random variables with common unknown mean θ. The variance
covariance matrix V of thee vector ( , ) , is such that the inverse of V has all its
diagonal elements equal to c and all its off-diagonal elements equal to d. Let be the best
linear unbiased estimator of θ and be the ordinary least squares estimator of θ. Which of
the following statements are true?
(a) = ∑ = .
(b) = n ̅ and = ∑ ̅ where ̅ is the mean of the ,s
(c) There are exactly (n 1) linearly independent linear functions of , each with
zero expectation.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page78

 (d) There are exactly (n 2) linearly independent linear functions of , each with
zero expectation.
Ans: a, c
27. Consider an M/M/1 queue with arrivals as a Poison at a rate of 8 per hour and a service time
which is exponentially distributed at a rate of 6 minutes per customer. The waiting time of a
customer in the queue
(a) has a gamma distribution with p.d.f
( )
f (x) = {
(b) has distribution function given by
( )
f (x) = {
(c) has mean 4 minutes
(d) has mean 24 minutes
Ans: b, d
28. Let X be a 4 1 random vector with multivariate normal distribution with mean μ and
dispersion matrix ∑. Suppose, the eigenvalues of ∑ are = 6, = 3, = 2, = 1. Let
, , be the four principal components. Which of the following statements are correct
(a) The percentage of variation explained by the first two components is 95%
(b) The percentage of variable explained by the first three components is 95%
(c) , , are independent
(d) , , have distribution.
Ans: a, b, c
29. Consider a region R, which is a triangle with vertices (0, 0), (0, θ), (θ, 0), where θ > 0. A
sample of size n is selected at random from this region R. Denote the sample as {( ): i =
1, 2, …, n}. Then denoting ( ) = max ( , ) and ( ) ( , ), which
of the following statements are true?
(a) ( ) and ( ) and independent
( ) ( )
(b) MLE of θ is
(c) MLE of θ is ( )
(d) MLE of θ is max{ ( ) ( ) }
Ans: c
30. Let X = ( , ) be 4 1 random vector such that X ~ (0, ∑) where

∑=( )

is positive definite. Then, which of the following statements are true?

(a) and have identical distribution.
( )
(b) ~ .
( )
(c) {( ) ( ) }. ( )
~ .
( )
(d) ~ Ans: a, c, d
( )

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page79

CSIR NET June-2015
Single Choice Questions [SCQ]
1. {N(t): t 0} is a Poisson process with rate > 0. Let = N(n), n = 0, 1, 2, …. Which of
the following is correct?
(a) { } is a transient Markov chain.
(b) { } is a recurrent Markov chain, but has no stationary distribution.
(c) { } has a stationary distribution.
(d) { } is an irreducible Markov chain.
Ans: a
2. Ten balls are put in 6 slots at random. Then the expected total number of balls in the two
extreme slots is
(a) 10/6
(b) 10/3
(c) 1/6
(d) 6/10
Ans: b
3. Suppose , X are random variable such than converges in distribution to X and
( ) also converges in distribution to X. Then
(a) X must have a symmetric distribution.
(b) X must be 0.
(c) X must have a density.
(d) must be a constant.
Ans: a
4. Assume that X ~ Binomial (n, p) for some n 1 and 0 < p < 1 and Y ~ Poisson (λ) for some
λ > 0. Suppose E [X] = E [Y].
Then
(a) Var (X) = Var (Y)
(b) Var (X) < Var (Y)
(c) Var (Y) < Var (X)
(d) Var (X) may be larger or smaller than Var (Y) depending on the values of n, p and λ
Ans: b
5. Let X, Y be independent random variables and let Z = + 3. If X has characteristic
function and Y has characteristic function , then Z has characteristic function θ where
(a) θ(t) = (2t) (-2t).
(b) θ(t) = ( ) ( ).
(c) θ(t) = ( ) ( ).
(d) θ(t) = ( ) ( ).
6. Let be a random sample from N (μ, ) where μ, and are unknown.
Consider the problem of testing : μ > 2. Suppose the observed values of are
1.2, 1.3, 1.7, 1.8, 2.1, 2.3, 2.7. If we use the Uniformly Most Powerful Test, which of the
following is true?
(a) is accepted both a 5% and 1% levels of significance.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page80

 (b) is rejected both a 5% and 1% levels of significance.
(c) is rejected at 5% levels of significance, but accepted at 1% levels of significance.
(d) is rejected at 1% level of significance, but accepted at 5% level of significance
Ans: a
7. Consider the model = iβ + , i = 1, 2, 3 where are independent with mean 0 and
variance respectively. Which of the following is the best linear unbiased
estimate of β?
(a)
(b) ( )
(c)
(d)
Ans: c
8. Suppose ~N( ), i = 1, 2 are independently distribution and are i.i.d N
( ), where , and are known. Then which of the following is true about the
marginal distributions of ?
(a) are i.i.d N ( ).
(b) are not normally distributed.
(c) are N ( ). but they are not independent.
(d) are normally distributed but are not identically distributed.
Ans: a
9. Let Y = ( )’ have the multivariate nomal distribution (0, I). Which of the following
is the covariance matrix of the conditional distribution of Y given
∑ ? (1 denotes the n vector with all elements 1.)
(a) I
(b) I + .
(c) I
(d)
Ans: c
10. Let be independent and identically distributed random variables having an
exponential distribution with mean . Let = and N = inf { n 1:
> 1 }. Then Var(N) equals
(a) 1
(b)
(c)
(d)
Ans: b
11. Suppose there are k strata on N = kM units each with size M. Draw a sample of size with
replacement from the stratum and denote by ̅ the sample mean of the study variable
∑ ̅
selected in the stratum i =1, 2, …, k define ̅ = ∑ ̅ and ̅ = .
Which of the following is necessarily true?

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page81

 (a) ̅ is unbiased but ̅ is not unbiased for the population mean.
(b) ̅ is not unbiased but ̅ is unbiased for the population mean.
(c) Both ̅ and ̅ are unbiased for the population mean.
(d) Neither ̅ nor ̅ is unbiased for the population mean.
Ans: a
12. Consider a Balanced Incomplete Block Design (BIBD) with parameters (b, k, v, r, λ). Which
of the following cannot possibly be the parameters of BIBD?
(a) (b – 1, k – λ, b – k, k, λ).
(b) (b, v – k, v, b – r, b – 2r +λ).
( )
(c) ( ).
(d) ( ).
Multiple Choice Questions [MCQ]
13. Let (X, Y) have the joint discrete distribution such that X|Y = y ~ Binomial (y, 0.5) and Y ~
Poisson (λ) λ > 0, where λ is an unknown parameter. Let T = T(X, Y) be any unbiased
estimator of λ then
(a) Var (T) Var (Y) for all λ
(b) Var (T) Var (Y) for all λ
(c) Var (T) for all λ
(d) Var (T) Var (Y) for all λ
Ans: b, c
14. Let be independent and identically distributed normal random variables with mean
0 and varance 1. Let be independent and identically distributed U (0, 1) random
variables, independent of . Define Z = . Then,
√

(a) E (Z) = 0.
(b) Var (Z) = 1.
(c) Z is standard Cauchy.
(d) Z ~ N (0,1)
15. Consider a Marco chain with state space S = {0, 1, 2, 3} and with transition probability
matrix P given by

P=

( )
Then
(a) 1 is a recurrent state.
(b) 0 is a recurrent state.
(c) 2 is a recurrent state.
(d) 3 is a recurrent state.
16. Consider a sample of size one, say X, from a population with pdf
f(x) = (x – θ) θ x 2θ, θ > 0
=0 otherwise

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page82

 Which of the following is/are confidence interval(s) for θ with confidence coefficient 1 – α?
(a) [ ]
√

(b) [ ]
√ √

(c) [ ]
√

(d) [ ]
√ √

Ans: a, b, c, d
17. Let X and Y be random variables with joint cumulative distribution function F (x, y). Then
which of the following conditions are sufficient for (x, y) to be a point of continuity of
F?
(a) P(X = x, Y = y) = 0.
(b) Either P (X = x) = 0 or P(Y = y) = 0.
(c) P ( X = x ) = 0 and P (Y = y) = 0
(d) P (X = x, Y y) = 0 and P (X x, Y = y) = 0.
Ans: c, d
18. Let X and Y be independent normal random variable with mean 0 and variable 1. Let the
characteristic faction of XY be denoted by . Then
(a) (2) = 1/2.
(b) is an even function.
(c) (t) = | | for all t 0.
(d) (t) = E ( ).
19. Suppose ,… are independent random variables. Assume that ,… are identically
distributed with mean and variance , while , … are identically distributed with
mean variable . Let = + … + . Then converges in distribution of
N (0, 1) if
( )
(a) = and =√ √ .
( ) ( )
(b) = and =
( )
(c) = ( ) and √

(d) = ( ) and √ √
Ans: a
20. Suppose X has density f (x|θ) = , x > 0 where θ > 0 is unknown. Define Y as follows:
Y = k if k X < k +1, k = 0, 1, 2, … then the distribution of Y is
(a) Normal
(b) Binomial
(c) Poisson

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page83

 (d) Geometric
Ans: d
21. To check whether a premium version of petrol given better fuel efficiency, a random sample
of 10 cars of a single model were tested with both premium and stared petrol. Let the
mileages obtained be denoted by ( ),… ( ), where denotes the mileage from
standard and from the premium form the car. We want ot test
: There is no difference in fuel efficiency between the two versions of petrol
Versus
: Premium petrol gives better fuel efficiency
Let = , ̅ = ̅ - ̅, = Rank of when | | are ordered. It is felt that fuel
efficiency measurements are not normally distributed and hence a nonparametric test is to be
proposed. Then which of the following can be considered suitable statistics for this purpose?
(a) ̅ - ̅ .
(b) Number of positive ’s
(c) Sum of corresponding to positive .
̅
(d) ̅)
√∑(
Ans: b, d
22. Consider the pdf f(x; θ, ) = ( ) + 0.1 (x ), where < θ < and > 0 are
of unknown parameters and denotes the pdf of N (0, 1). Let be a random
sample form this probability distribution. Then which of the following is (are) correct?
(a) This model is not parametric.
(b) Method of moments for and exist.
(c) An unbiased estimator of exist.
(d) Consistent estimators of θ do not exist.
Ans: b, d
23. Suppose X has density f(x|θ) where θ is 0 or 1. Also, f (x|0) = 1 if 0 < x < 1 and 0 otherwise,
f (x|1) = if 0 < x < 1 and 0 otherwise. To test : θ = 0 versus : θ = 1 at level α, 0 <
√
α < 1 , the Most Powerful test
(a) reject if X > 1 – α
(b) reject if X < α
(c) reject if X < √
(d) has power √
Ans: b, d
24. Consider the linear model Y ~ (Xβ, ), where X is a n (k + 1) matrix of rank k + 1 <
̂
n. let and ̂ be the maximum likelihood estimators of β and respectively. Then which
of the following statements are true?
(a) Cov ( ̂ ) =
(b) ̂ and ̂ are independently distributed
(c) ̂ is sufficient for
(d) ̂ where A is a suitable matrix of rank (n – k - 1).
Ans: b, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page84

25. Let be uncorrelated observation with common variable and expectations given
by E( ) = + , E( ) = + , E( ) = + , where s are unknown parameters.
In the framework of the linear model which of the following statement(s) is (are) true?
(a) Each of , and is individually estimable.
(b) ∑ is estimable
(c) and are each estimable.
(d) The error sum of squares is zero.
Ans: c, d
26. Consider a random sample ( ),…,( ) from the bivariate normal distribution with
E( ) = μ = E( ), Var( ) and Cov( ) = for all i. Let ̂ ̂ and ̂ denote the
maximum likelihood estimators of μ, and respectively. Also,
=∑ ( ̅) , =∑ ( ̅ ) and ∑ ( ̅ )( ̅ ), then
(a) and are independent.
(b) ̂ (̅ ̅ ), ̂ = ( ), ̂ = .
(c) ̂ (1 + ̂) = ( ).
(d) ̂ (1 + ̂) = ( ).
Ans: a, b, c, d
27. Consider the linear model
=
=

=
=
Where ,… , are unknown parameters and , … , are uncorrelated with mean 0
and common variable. Let Y be the column vector ( ) and ̅ = ∑ . Which of
the following are correct?
(a) If E ( Y) = 0, then all elements of are equal.
(b) The best linear unbiased estimator of .
(c) The best linear unbiased estimator of ̅
(d) All linear functions + …+ are estimable.
Ans: a, d
28. Let X (t) = number of customers in the system at time t in as M/M/C queuing model, with C
= 3, arrival rate λ > 0 and service rate μ > 0. Which of the following is/are true?
(a) {X (t)} is a birth and death process with constant birth and death rates.
(b) If {X (t)} has a stationary distribution, then λ < 3μ.
(c) If λ>3μ, then then thee stationary distribution is a geometric diminution with
parameter .
(d) The number of customers undergoing service at time t is min {X (t), 3}.
29. Let ∑ = (( )) be an n n symmetric and positive be finite matrix such that 0 for all i,
j. Which of the following matrices will always be the covariance matrix of a multivariate
normal random vector?
(a) ∑.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page85

 (b) The matrix with the (i, j)th element for each i, j.
(c) The matrix with (i, j)th element for each i, j
(d) ∑
Ans: a, b, d
30. Let be i.i.d. standard normal variable. Which of the following has wwishart
distribution with 2 d.f ?
(a) [ ]

(b) [ ]

(c) [ ]

(d) [ ]
Ans: d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page86

CSIR NET Dec-2014
Single Choice Questions [SCQ]
1. Suppose is a random sample from U (0, θ). θ > 0. Let ( ) ( ) … ( )

be the order statistics. Consider the two unbiased estimators for θ: = 2 ̅ and =( )
( ) . Then
( )
=
( )
(a) 0
(b) 1
(c)
(d) 12
Ans: a
2. Suppose than are independent and identically distributed random variable each
having a Bernoulli distribution with parameter ½. Consider the 2 2 matrix A =( ).
Then P (A is invertible) equals
(a) 0
(b) 1
(c) 1/3
(d) ¼
Ans: c
3. Consider the linear programming problem:
Minimize: z =
Subject to 3x + 4y 5, x 0, y 0.Which of the following is correct?
(a) Set of feasible solutions is empty.
(b) Set of feasible solutions is non-empty but there is no optimal solution
(c) Optimal value is attained solution at( ).
(d) Optimal value is attained solution at( ).
Ans: b
4. How many distinct samples of size n can be drawn with replacement from the population
{ } of n units?
(a)
(b) ( )

(c) ( )
(d) 1
Ans: b
5. For testing the effectiveness of four teaching techniques, five teachers of a college were
involved. A call of 120 students was divided into 5 groups of 24 each at random; one group
was assigned to each of the five teachers. Each teacher further divided his/her group into four
equal subgroups at random, and used one technique per subgroup. All of them used the same
course material. After all the class were over, a common examination was conducted and the

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page87

 marks were noted. Suppose we want to test whether all the four teaching techniques are
equally effective. What are the degrees of freedom associated with the residual sum squares?
(a) 60
(b) 100
(c) 119
(d) 460
Ans: b
6. Suppose … are random variables on a common probability space with ~ N
( ). Then, converges in probability to 0 if and only if
(a)
(b)
(c) converges
(d) converges
Ans: b
7. Let … be independent and identically distributed random variables with E ( ) = 0
and Var( ) = 1 for all i. Let = Let ϕ (x) denote the cumulative distribution
function of a standard normal random variable. Then , for any x > 0, (
) equals
(a) 2ϕ(x) – 1
(b) ϕ(x)
(c) 1
(d) 1 - Φ(2x)
8. Let … be a random sample of size n from a p-vector Normal distribution with mean
and positive definite covariance matrix ∑. Choose the correct statement.
(a) ( ) ∑ ( ) has chi-square distribution with 1 d.f.
(b) ̅ ̅ has Wisher distribution with p d.f.
(c) ∑ ( )( ) was Wisher distribution with n d.f
(d) and are independently distributed.
9. Suppose X~ N(0, 1) and the conditional distribution of Y given X = x is N ( x, 1), for 0 <
< 1. When we regress Y on X, the coefficient of determination is
(a)
(b)
(c) √
(d)
10. Let X and Y be integer-valued, bounded random variables. Then which of the following
statement is FALSE?
(a) E (X) = ∑ ( ) ( )
(b) V (X) = ∑ ( ) ( )
(c) P (X = x) = ∑ ( ) ( )
(d) E (XY) = ∑ ( ) ( )
11. Suppose X ~ Poisson (λ), λ > 0. Let the prior distribute of λ be U (0, 1). If X = 0 is observed,
then the posterior probability of the set 0 < λ is
(a) 0.5

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page88

 (b) 1
√
(c)
(d)
Ans: c
12. Five persons A, B, C, D and E are seated at random of eight numbered chairs which are
arranged in a circle. What is the probability that A and B are separated by at least 2 chairs?
(a) 3/7
(b) ½
(c) 4/7
(d) ¼
Ans: a
Multiple Choice Questions [MCQ]
13. Let , …, n 3, be a random sample from N ( ) population where is unknown.
Define ̅ = ∑ . Which of the following are necessarily true?
(a) Cov ( ̅ )=0
(b) Cramer-Rao lower bound for unbiased estimators of is

(c) Var( ̅ ) < Var ( ( ) )

(d) ̅ is a function of any sufficient statistic for μ
Ans: a, b, c, d
14. Consider a Markov chain with state space {1, 2, …,100}. Suppose states 2i and 2j
communicate with each other and states 2i – 1 and 2j – 1 communicate with each other for
( ) ( ) ( )
every i, j = 1,2,…,50. Further suppose that > 0, > 0 and > 0. Then
(a) The Markov chain is irreducible.
(b) The Markov chain is periodic.
(c) State 8 is recurrent.
(d) State 9 is recurrent.
Ans: c
15. Let , …, be independent and identically distributed observations from the
distribution with density f ( | ) = ( ) , x > where < < and is unknown. Let
=2 ∑ and ( ) = 2 ( ) where ( ) is the smallest order statistic. To test : > 0 at
level α, where 0 < α < 1. Consider the two tests A and B given below.
A. Reject if > where is such that P ( ) = α with ~
B. Reject if > where is such that P ( ) = α with ~
The which of the following statements are valid?
(a) Both A and B are level α tests
(b) A is the uniformly most powerful level α tests
(c) B is the uniformly most powerful level α tests
(d) B is more powerful than A at any > 0
Ans: a, c, d
16. Let , …, be independent random variables with E ( ) = 0 and Var ( ) = k. Let
=∑ . Then, as n ,

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page89

 (a) 0 in probability
(b) 0 in distribution
(c) 0 in distribution
(d) 0 in probability
Ans: a, b, c, d
17. Let ϕ (t) be a characteristic function of some random variable. Then, which of the following
are also characteristic function?
(a) f (t) = [ ( )] for all t
(b) f (t) = ( ) for all t
(c) f (t) = ( ) for all t
(d) f (t) = ( ) for all t
Ans: a, b, c
18. Suppose Let , and are independent and identically distributed abloom variables,
having density function f. then,
(a) P( ( ) )=
(b) P( ( ) )=
(c) P( ( )) =
(d) P( ( ))) =
Ans: a, c
19. N, … are independent real valued random variables such that P(N = k) = (1 p) , k
= 0, 1, 2 … where 0 < p < 1, and { } is a sequence of independent and
identically distributed bounded random variables. Let
( )
X (w) = {
∑ ( )
Which of the following are necessarily correct?
(a) X is a bounded random variable
( )
(b) Moment generating function of X is (t) = ,t ,
( )
Where is the moment generating function of .
(c) Characteristic function of X is
( )
(t) = ,t , where is the characteristic function of .
( )
(d) X is symmetric about 0.
Ans: c
20. Suppose X is a random variable such that E (X) = 0, E( ) = 2 and E( ) = 4. then
(a) E( ) = 0
(b) P(X ) = 1/2
(c) X ~ N(0, 2)
(d) X is bounded with probability 1.
Ans: a, c, d

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page90

21. Consider a 3- variate population with Covariance matrix ( )

Where > 0, > 0. Then which of the following statements are true?
(a) <
√
(b) The proportion of the total population variance explained by the first principal component
is
(c) The second principal component is uncorrelated with the first and the third principal
component.
√
(d) The proportion of the total population variance explained by the first ( + √ )
Ans: a, c, d
22. A and B are two methods to determine the levels of mercury in fish. In a study to compare A
and B amount of mercury was measured using both methods on n = 12 fish. Let ( ), …,
( ) be those measurements, with ’s standing for method A and ’s for method B. It
should be boted that the size of error in measurement can depend on the amount of mercury,
so the observations ( ), …, ( ) may not be identically distributed. To test : There
is no difference between methods A and B versus : Method B typically gives a larger
reading than method A, which of the following test statistic are appropriate?
(a) Number of pairs ( ) with ( ), 1 i n; 1 j n
(b) Sum of the ranks of the Y observations in the combined sample
(c) Number of the pairs ( ) with ( ), 1 i n
(d) ̅ ̅
Ans: c
23. Let Y follow multivariate normal distribution (0, I) and let A and B be n n symmetric,
idempotent matrices. Then which of the following statements are true?
(a) If AB = 0, then and are independently distributed.
(b) If ( ) has chi-square distribution then and are independently
distributed.
(c) ( ) has chi-square distribution
(d) and have chi-square distribute
Ans: a, b, d
24. For a Markov chain with finite state space, the number of stationary distributions can be
(a) 0
(b) 1
(c) 2
(d)
Ans: b, d
25. Let X (t) = number of customers at time t in the system in an M/M/1 queuing model with
arrival rate λ > 0 and service rate μ > 0. Let = ( ( )) = k), k = 0, 1, 2 …
whenever it exists. Which of the following are true?
(a) {X(t) } is a birth and death process with birth rates = λ, k = 0, 1, 2, … and death rates
= , k = 1, 2, …

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page91

 (b) {X(t)} is a birth and death process with birth rates = , k = 0, 1, 2, … and death rates
= , k= 1, 2, …
(c) Limiting distribution { } exists if and only if > , and is the geometric with parameter
( )
(d) If an arriving customer finds exactly one costumer, then his total waiting time in the
system has an exponential distribution with parameter (2 )
Ans: a, c
26. Consider the linear model
= 2θ + β +
=β+ +
=θ+β+ +
Where θ, β, are unknown parameters and , , are uncorrelated random errors whit
mean 0 and constant variance. Then which of the following statements are true?
(a) θ, β and are estimable
(b) is estimable
(c) 2θ is estimable
(d) is estimable
Ans: b, c
27. Suppose a sample of size n is drawn using simple random sampling without replacement
from a finite population of N units where N > n and denote the sample mean of the study
variables corresponding to in unit and draw a simple random sample of size n without
replacement from the remaining (N 1) units and denote the sample mean of the study
variables corresponding to the selected units by ̅ . Define =N ̅ = (N 1) ̅ + ,
= Var ( ) and = Var ( ). Which of the following are necessarily true?
(a) is unbiased for population total
(b) is unbiased for population total
(c) = where = population variance
(d) for all n, N
Ans: a, b, c, d
28. Consider the linear model E(y) = Xβ, Cov(Y) = I where X is a matrix of size n p having
rank r p. Then which of the following statements are necessarily true?
(a) The set of estimable linear functions from a vector space of dimension r.
(b) If E ( Y) = 0 for some nonzero vector c, then there is a function β which is not
estimable.
(c) If all linear functions β are estimable, then r = p
(d) The set of function Y with E( Y) = 0 form a vector space of dimension r.

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page92

29. Consider a factoral experiment with three factors A,B and C. suppose eight treatments are
assigned in two blocks of each of the four replicates in the following way.
Replicate 1 Replicate 2 Replicate 3 Replicate 4
(1) b (1) a (1) a (1) b
a c b c c b ab a
bc ac ac bc ab ac bc c
abc ab abc ab abc bc ac abc

Which of the following are necessarily true?

(a) This is an example of complete confounding
(b) AB is confounded in Replicate 1
(c) AC is confounded in Replicate 2
(d) ABC is confounded in Replicate 4
Ans: c, d
30. Let , …, be independent and identically distributed Bernoulli where 0 < θ < 1 and n
> 1. Let the prior density of θ be proportional to , 0 < θ < 1. Define S = ∑ .
√ ( )
Then valid statements among the following are:
(a) The posterior mean of θ does not exist
(b) The posterior mean of θ exist
(c) The posterior mean of does not exist it is larger than the maximum likelihood estimator
for all values of S
(d) The posterior mean of θ exists and it is larger than the maximum likelihood estimator for
some values of S.
Ans: b, d

Visit us: http://pstatisticstutorials.com/

Youtube: PStatisticsTutorials
Telegram Channel: https://t.me/Pstatisticstutorials

PST|CSIR NET|GATE|IIT-JAM|ISS |BY-PANKAJ SIR|Mob-8574270238|Website:pstatisticstutorials.com|Page93