Scribd
Upload
44 views2 pages

JRF Stat STA 2019

1. The document contains 10 problems related to statistics and probability. It provides definitions, formulas, and questions regarding topics like sequences, limits, probability distributions, random variables, estimators, and experimental design. 2. The problems involve calculating limits, determining distributions of transformed random variables, identifying maximum likelihood estimators, analyzing covariance matrices, suggesting models to test beliefs about hospital admission patterns, and assessing properties of randomized block designs. 3. Methods employed include determining limits of sequences and integrals, deriving joint distributions, solving likelihood equations, expressing covariance matrices in terms of other matrices, specifying assumptions for hypothesis testing, and evaluating properties like equireplication, connectedness and orthogonality for experimental designs.

Uploaded by

Yojan Sarkar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
44 views2 pages

JRF Stat STA 2019

1. The document contains 10 problems related to statistics and probability. It provides definitions, formulas, and questions regarding topics like sequences, limits, probability distributions, random variables, estimators, and experimental design. 2. The problems involve calculating limits, determining distributions of transformed random variables, identifying maximum likelihood estimators, analyzing covariance matrices, suggesting models to test beliefs about hospital admission patterns, and assessing properties of randomized block designs. 3. Methods employed include determining limits of sequences and integrals, deriving joint distributions, solving likelihood equations, expressing covariance matrices in terms of other matrices, specifying assumptions for hypothesis testing, and evaluating properties like equireplication, connectedness and orthogonality for experimental designs.

Uploaded by

Yojan Sarkar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

STA

2019

1. (a) Let {xn }n≥1 be a sequence of real numbers converging to a finite real
number a, as n → ∞. Define
 1
 xn − n if n = 3k,
yn = 2xn if n = 3k − 1,
 3xn +1
3|xn |+1 if n = 3k − 2,

for k = 1, 2, . . .. Find the values of a for which the sequence has three dis-
tinct limit points.
�1 √
(b) Let an = 0 (1 − x2 )n dx for n ≥ 1. Find lim n an . Also show that
n→∞
�
∞
an diverges. [6+6]=12
n=1

2. Let f be a real-valued, continuous, strictly increasing function defined on


[0, 1] with f (0) = 0 and f (1) = 1. Let g be the inverse function of f . Prove
that � �
1 1
f (x)dx + g(y)dy = 1.
0 0 [12]

3. (a) Let x be the n × 1 vector with xi = i for i = 1, . . . , n. Find the


determinant of I + xxT , where I is the identity matrix of order n.
(b) Let A and G be matrices of order m × n and n × m, respectively, such
that AGA = A. Show that the determinant of I + AG is non-zero.
[6+6]=12

4. Suppose that two distinct positive integers are chosen randomly from 1
to 50. What is the probability that their difference is divisible by 3?
[12]

5. Consider a bivariate random vector (X1 , X2 ) having joint density given


by �
4x1 x2 , if 0 ≤ x1 ≤ 1, 0 ≤ x2 ≤ 1,
f (x1 , x2 ) =
0, otherwise.
Define two new random variables as Y1 = X1 /X2 and Y2 = X1 X2 .

a) Find the support of the joint distribution of (Y1 , Y2 ) and represent it


graphically.

b) Derive the joint distribution of (Y1 , Y2 ).

c) Using (b), or otherwise, calculate P (Y1 > 2, Y2 > 1/4).


[4+3+5]=12

1
6. Suppose X1 , X2 , . . . are independent and identically distributed random
variables with P (Xi = 1) = 14 = P (Xi = −1) and P (Xi = 0) = 12 for all
�n
i = 1, 2, . . .. Define Sn = Xi and Un = sgn(Sn ), for n ≥ 1, where the sgn
i=1
function is given by
�
1, if x ≥ 0,
sgn(x) =
−1, if x < 0.
Find lim P (Un ≤ u) for u ∈ R, and hence identify the limiting distribution
n→∞
of Un . [12]

7. Let X1 , X2 , . . . , Xn be independent and identically distributed having


a common density fθ where θ ∈ Θ ⊆ R. Let θ� be the unique maximum
likelihood estimator (MLE) of θ; note that θ� solves the likelihood equation.
Let T (X1 , . . . , Xn ) be an efficient estimator of τ (θ), a one-to-one function of
the parameter θ, in the sense that T is unbiased for τ (θ) and its variance
attains the corresponding Cramer-Rao lower bound. Prove that T must be
an MLE of τ (θ). [12]

8. Suppose that a n-variate random vector X = (X1 , X2 , . . . , Xn )T ∼


Nn (µ, In − n1 Jn ), where Jn is n × n matrix with each entry 1. Define the
multivariate log-normal random vector Y = (Y1 , Y2 , . . . , Yn )T through the
relations Xk = log Yk for k = 1, . . . , n. Prove that the covariance matrix of
Y can be expressed as DBD for some n × n matrix B and some diagonal
matrix D. Find these matrices. [12]

9. It is believed that the number of daily hospital admissions on the average


depends on whether it is in weekdays or in weekends. Based on records
of daily admission counts from a hospital over 52 weeks, suggest a suitable
model and corresponding analysis to test this belief. Write your notation
and all the assumptions clearly, including limitation(s), if any.
[10+2]=12

10. Consider a randomized block design with v treatments and b blocks


with v < b. But, suppose that observations under treatment i in block i are
missing for i = 1, 2, . . . , v, and the resulting design is denoted by D.
a) What are block sizes in D? Are the treatments equi-replicated in D?
b) Is the design D connected? Justify your answer.
c) Using (b), or otherwise, prove that the design D is not orthogonal.
[2+4+6]=12

You might also like