ASSIGNMENT-12

Statistical inference

 n2 
S 
(1.) Let X 1 , X 2 ,..., X n be i.i.d. Poisson with mean 1. Using CLT ' x ' in terms of t if P  n 2     x
 t
 n 
n
where Sn   X i .
i 1

x  n  t  1  
n
(a)
 2

x  n  t  1  
n
(b)
 2

x  n  t  1  
n
(c)
 2

x  n  t  1  
n
(d)
 2
Ans: (a)

 40 
Let X i ~ bernoulli   for i  1,..., 40 . Define Yi  2 X i3  X i for each i then the value of P   Yi  25 
1
(2.)
2  i 1 
equals
(a) 1   1.42 
(b) 1   1.50 
(c) 1   1.64 
(d) 1   1.58 
Ans: (d)

(3.) Let 3 i.i.d. sample be taken from A ~ N 1,3 and 4 i.i.d. samples be taken from B ~ N  3, 4  , A and B
are independent. Let S12 and S22 denote the sample variance. Then the prob. that S22  3S12 is _____
(a) F3,2  2 
(b) F3,2  3
(c) F2,3  2 
(d) F2,3  3
Ans: (a)

(4.) A sample of size 6 is taken from a normal population variance is estimated using sample variance.
 Then the prob. that the absolute relative error of estimation is less than 0.54 equal
(a) 0.975
(b) 0.1
(c) 0.95
(d) 0.9
Ans: (d)

(5.) Consider two independent population A and B distributed as N  60, 4  and N  80,3 respectively.
Suppose we take a sample consisting of size 4 from A and 3 from B . Then the prob. that the total
sample mean is less than 70 equal?
(a)   3
(b)   2
(c)   0.2 
(d)   0.3
Ans: (b)

(6.) Let X 1 , X 2 , X 3 ~ iidN  0,  2  random variable if k  max  X 1 , X 2 , X 3   min  X 1 , X 2 , X 3   is unbiased

estimator of  then the value of k equals.

(a)
3

(b)
2

(c)
6

(d)
9
Ans: (a)

(7.) Consider n sample X 1 , X 2 , X 3 ,..., X n i.i.d. sampled from Ber   . Given X denote the sample mean.
Then an unbiased estimator for  1    is
n
(a) X
n 1
n 1
(b) X 1  X 
n
(c) X 1  X 
1
(d) X 1  X 
n 1
 Ans:

(8.) Consider X ~ U   ,  ,   0 if define Y   log X . Let Y1 , Y2 ,..., Yn be n random sample from pdf
of Y say f Y  then which of the following will be a unbiased estimator of  .
n 1
(a)
Y
1
(b)
Y
n 1
(c)
nY
n
(d)
 n  1 Y
Ans: (c)

(9.) iven samples of size 3, 4 are taken from independent populations following distribution N 1,12 
Given
and N  0.5,4  . Let X 1 and X 2 denote the respectively sample means. What is the prob. that X 1  2 X 2
?
(a) 0.15
(b) 0.5
(c) 0.25
(d) 0.3
Ans: (c)

(10.) 10 i.i.d. sample are drawn from a Cauchy distribution with pdf
1
f  x  for x  R f r  x  is pdf of r th order statistic, then which of following is true?
  x  3 2

(a) f 2 1  f 9  5
(b) f 2  3  f 9  5 
(c) f 3 1  f8  5
(d) f 3  3  f 8  5 
Ans: (a)

(11.) Let X 1 , X 2 ,..., X n be n i.i.d sample drawn from a population distribution as N   ,  2  . Let T1 be the
sample variance when  is unknown and T2 be the sample variance when  is known. If both are
being used to estimate  2 then:
(a) T1 is biased , T2 is biased
 (b) T1 is biased , T2 is unbiased
(c) T1 is unbiased , T2 is unbiased
biased
(d) T1 is unbiased , T2 is biased
Ans: (b)

(12.) Let r sets of sample of size n1 , n2 ,...nr be take n from a population N   ,  2  . Let T be an unbiased
estimator for  2 such that
r ni

n s  X  Xi 
2 2
i i i
T i 1
where Si2  i 1
then value of a equal
a ni

(a) n1  n2  ...  nr  r
(b) n1  n2  ...  nr
(c) n1  n2  ...  nr  r
1
(d)  n1  n2  ...  nr    r 
2
Ans: (c)

(13.) Let X 1 , X 2 ,..., X n be n i.i.d. samples drawn from a population distributed as N   ,  2  where  is
known, then which of the following is an unbiased estimator for  ture ?
 n
(a) T
n
X
i 1
i 

 n
(b) T
2n
X i 1
i 

2 n
(c) T
n
X
i 1
i 

 n
(d) T n 1  X i  
2 i 1

Ans: (d)

(14.) Let X 1 , X 2 ,..., X n be n i.i.d. samples drawn from a population distributed as Exp    . Consider the
n
estimator T  n
for estimating  . Which of the following is True about T ?
 Xi
i 1

(a) unbiased and consistent

(b) biased and not consistent
(c) biased and consistent
 (d) unbiased and not consistent
Ans: (c)

(15.) Let X 1 , X 2 ,..., X n be n i.i.d sample from a population distributed as N   ,  2  . Let X denote the
denote mean then which of the following is TRUE?
(a)  X  1 X  2   X  3 is consistent estimator for  3  6 2  11  6
(b)  X  1 X  2   X  3 is consistent estimator for  3  6 2  11  6
(c)  X  1 X  2   X  3 is consistent estimator for  3  6 2  11  6
(d)  X  1 X  2   X  3 is consistent estimator for  3  6 2  11  6
Ans: (b)

(16.) Let T  X an estimator for  such that the sample are drawn independently from N   ,  2  where  2
is known. Then which of following is/are TRUE?
n 1
(a) I   
2
n
(b) I   
2
(c) there exist another unbiased estimator T  , such that var T    var T 
(d) there does not exist another unbiased estimator such that var T    var T 
Ans: (b), (d)

(17.) Let T1 be MVUE for  and T2 be another unbiased estimator with efficiency 0.25, then coefficient of
correlation between T1 and T2 equals
(a) 0.40
(b) 0.50
(c) 0.30
(d) 0.64
Ans: (b)

(18.) Consider T  X to be an estimator for  when X 1 , X 2 ,..., X n be n i.i.d sample drawn from a
population distributed as P    , then which of the following is/are TRUE?
(a) I  log   is independent
ndependent of 
(b) I  log   is depend on 

(c) T T  1T  2  is UMVUE for  3

T T  1T  2 
(d) 3
is UMVUE for  3
n
Ans: (c)
(19.) C  R lower bound theorem cannot be applied for parametric estimator of  for which of the
following?
(a) Bin  ,0.01 

(b) Beta  , 

(c) Gamma  2 , 
(d) Poi  
Ans: (a)

(20.) Let X 1 , X 2 ,..., X n be i.i.d samples drawn from a population distributed as N  ,1 then the UMVUE
for  2 is
1
(a) X2 
n
(b) X2
X2
(c)
n
1
(d) X2 
n
Ans:

(21.) Let T be an estimator of a parameter  of a population with distribution f  x  .Consider samples

X 1 , X 3 ,..., X n taken independently from the distribution such that
f  X 1 , X 2 ,..., X n   g  t  h  X 1 , X 2 ,..., X n  where g  t  is a function of the parameter  and the
 X 1  X 2  ...  X n if   0

estimator T . While h  X 1 , X 2 ,..., X n  is defined as h  X 1 , X 2 ,..., X n    X 1  X 2  ...  X n , then
 if   0
2
(a) T is sufficient for 
(b) T is NOT sufficient for 
(c) can’t say any thing about T
(d) T depends  n
Ans: (c)

(22.) Consider i.i.d sample X 1 , X 2 ,..., X n taken from N   ,  2  where  2 is know. Let X denotes the
sample mean then which of the following is TRUE?
(a) e X is sufficient for e 
(b)  X  1 X  2  is sufficient for    1   2 
 (c)  X  1 X  2  is sufficient for    1   2 
(d) None of the above
Ans: (a)

(23.) Consider a random variable X , with pdf f  X  for which of the following, three does not exist a
sufficient statistic (ignore trivial statistic)
2 x   
(a) f  x  , 0  x 
2
(b) 4e 4  x  , x  
1
(c) f  x  , x
 1   x   2
2x
f  x  /
, x0
2
(d) e x

Ans: (c)

(24.) Let X 1 , X 2 ,..., X n be random sample 1  a, b  then sufficient statistic for b equals
n
(a) x
i 1
i

n
(b) x
i 1
i

n
(c) n   xi
i 1

n
(d)  1  x 
i 1
i

Ans: (d)

(25.) Let X 1 , X 2 ,..., X n be random sample 1  a, b  then sufficient statistic for  equals
n
(a)  1  X 
i 1
i

n
(b)  X 1  X 
i 1
i i

n
(c) X
i 1
i

n
(d)  X 1  X 
i 1
i
2
i

Ans: (b)
(26.) Let X 1 , X 2 ,..., X n be random sample from a population having pdf

f  x   e  x , x  

then pdf of sufficient statistic for 

 ne  ny y0
(a) 
 0 otherwise

ne  n y   y0
(b) 
 0 otherwise

 1  y  
 e n y0
(c) n
 0 otherwise

 1  y/n
 e y0
(d) n
 0 otherwise
Ans: (b)

(27.) Let X 1 , X 2 ,..., X n i.i.d sample from a distribution gamma   ,  where both are unknown, then
sufficient statistic for   ,   equals?

(a)   X , X  jointly
i n

 n

(b)   X i ,  X i  jointly
 i 1 
 n

(c)   n   X i  jointly
X ,
 i 1 
(d)  X  n , X 1  jointly
Ans: (b)

(28.) mple are drawn from population with distribution on N  n ,0.1 and 5 samples are drawn from
10 sample
N  2 ,0.4  .For testing of hypothesis H 0 : 2  1 and H1 : 2  1 , H 0 : is rejected if X 1  X 2  0.6 . The
level of significance of test is equals
(a) 0.046
(b) 0.05
(c) 0.01
(d) 0.026
Ans: (a)
(29.) Let T be an unbiased estimator of population parameter  . Then which of the following is True?
(a) The CR bound always hold for the estimator T 2 while estimating  2 .
(b) The CR bound be not hold for the estimator T 2 while estimating  2 .
(c) The CR bound may be hold for the estimator T 2 while estimating  2 .
(d) CR bound depends on 
Ans:

(30.) Let X 1 , X 2 ,..., X n be i.i.d samples from a geometric distribution with mean 1 / p then the CRLB for
estimating p equals
1
(a)
p 1  p 

1
(b)
p 1  p 
2

1
(c)
p2
1
(d)
1 p
Ans: (b)

(31.) Consider n i.i.d sample X 1 , X 2 ,..., X n taken from a population with distribution p    . Then which of
the following statement(s) is/are TRUE?
n
 1
(a) e 1   X is consistent for 
 n
n
  Xi
(b) e is MLE for e  
(c) X  X  1 is sufficient for 
(d)  X  1 X  2  is sufficient for    1   2 
Ans:

(32.) If T  X   t  x1 , x2 , x3 ..., xn  is a sufficient estimator of parameter  , then the deriviative of log

likelihood with respect to  can be written
(a) lways as a function of only  always as a function of  if t  0
always
(b) always as a function of  if t  0
(c) always as a function of t if   0
(d) always as a function of t and 
Ans: (d)
(33.) Let X 1 , X 2 ,..., X n be i.i.d samples from distribution given as f  x     x   1 if x   ,   0,   0 if
 is known then which of the following statements is/are True?
(a) the MLE of  is min X i
nX 1  5
(b) is consistent for 
n 1
(c) log X 1 is sufficient for log 

(d) log X1 is MLE for log 

Ans: (a), (b), (c), (d)

(34.) Let Let X 1 , X 2 ,..., X n be i.i.d samples from U  ,  1 .

Define X  n  max X i , X 1  min X i , i  1, 2,..., n . Then which of the following statement is are True?
(a) MLE of  X  n 
(b) MLE of  not unique
(c) cov  X  n  , X  n   X 1   0

(d) X  n  is complete sufficient statistic for 

Ans: (c)

(35.) For a population with distribution 1  a, b  where let m1 and m2 represent the 1st and 2nd order
sample moment. The MOME for a, b equals
m2 m1  m12
(a) a
m12  m12

(b) b
 m1  1 m1  m2 
m12  m22

m2 m1  m12

(c) a
m12  m22

(d) b
 m1  1 m1  m2 
m12  m2
Ans: (b)

(36.) Let X 1 , X 2 ,..., X n be i.i.d samples from N  0,  2  . Define T  method of moments estimator of  2 ,
then which of the following statements is/are TRUE?

(a) T is unbiased
(b) T is less efficient than MLE estimator
(c) T is a function of sufficient statistics
 (d)
Ans: (a), (c), (d)

(37.) Consider 5 i.i.d sample from Exp    with H 0 :   0.5 and H1  0.1 then most powerful C.R. for
  0.05 equal
5
(a) X
i 1
i  18.31

5
(b) X
i 1
i  3.94

5
(c) X
i 1
i  3.94

5
(d) X
i 1
i  18.31

Ans: (a)

(38.) Which of the following statement is NOT TRUE?

(a) Roles of H 0 and H1 are symmetrical
(b) If testing H 0 :  0 and H1 :  1 then MPCR is uniformly most powerful C.R.
(c) Forr a given level of significance the critical region is not unique
(d) The MPCR reduces  and  simultaneously
Ans: (c)

(39.) Let X 1 , X 2 ,..., X n be i.i.d sample from a population with pdf f  x  parameterized by  . If T is a
sufficient statistic for  and g  t  denotes some function of T and  then MPCR can be written as
(a) W   X : g1  t   k  g0  t 

(b) W   X : g1  t   k  g0  t 

(c) W   X : g1  t   g0  t   k 

(d) W   X : g1  t   g0  t   k 

Ans: (a), (b)

(40.) Let X 1 , X 2 ,..., X n be i.i.d sample from N  0,  2  with H 0 :  0.5 and H1 :  1 which is NOT TRUE
about the geometry of most powerful critical region for   0.05 ?
(a) it is the region outside a sphere
(b) the sphere is centred at origin
(c) the radius of the sphere is approx 0.3
(d) the surface of the sphere is excluded
 Ans: (a)

(41.) Let X i , i  1, 2,..., n  1 be n n  1 i.i.d random variables such that X i ~ N  0,1 , i .

Xj  n 1 
let Yi  , E   Yi  is
n
 i 1 
X
i 1
i
2

i j
n
(a) 0
(b) n 1
1
(c)
n 1
1
(d)
1  n 2
Ans: (a)

 x 1 if 0  x  1
(42.) Let X 1 , X 2 ,..., X n be i.i.d sample from a population with pdf f  x    ,   0 . The
 0 otherwise
MPCR of significance  for testing H 0 :  1 against H1 :  2 is given by

 n

(a)  X : X i  exp  x1 ,2 n 
2

 i 1 

(b)
 1 2
 
 X : X i  exp  x1 ,2,2n 
 2 

(c)
 1 2

 X : X i  exp  x1 ,n, n 
 2 

(d)  X : X i  exp  x12 ,n 
Ans: (d)