LOYOLA COLLEGE (AUTONOMOUS), CHENNAI – 600 034
M.Sc. DEGREE EXAMINATION – STATISTICS
SECOND SEMESTER – APRIL 2012
ST 2811 / 2808 – ESTIMATION THEORY
Date : 17-04-2012 Dept. No. Max. : 100 Marks
Time : 9:00 – 12:00
SECTION – A
Answer all the Questions: (2×10=20 Marks)
- State the Methods of Obtaining UMVUE
- State the Invariance Property of MLE
- State Neyman-Fisher Factorization Theorem
- Provide an example to prove that an unbiased estimator need not be unique
- Define Sufficient Statistic and Provide an Example
- Define Bayesian Estimator
- State the use of Rao-Blackwell Theorem
- Define T-Optimality
- Provide the large sample behavior of Maximum Likelihood Estimator
- Define Best Linear Unbiased Estimator
SECTION – B
Answer any Five Questions: (5×8=40 Marks)
- State and Prove the necessary and sufficient condition for unbiased estimator to be UMVUE
- State and Prove Cramer-Rao Inequality for Multi-parameter case and hence
establish the inequality for the case of single parameter
- State and Prove Neyman-Fisher Factorization theorem
- Let X1,X2,…,Xn be a random sample of size n from uniform distribution U(0,θ),
Y=max{ X1,X2,…,Xn} show that is an Unbiased Estimator of θβ. Where β is a
positive constant
- State and Prove Rao-Blackwell Theorem.
- Let Y1,Y2,Y3,Y4 be random variable with E(Y1) = E(Y2)= θ1+ θ2 , E(Y3) = E(Y2)= θ1+ θ3 determine the estimability of the following linear parametric functions
- i) 2θ1+ θ2+ θ3 ii) θ3-θ2 iii) θ1 iv) 3θ1+ θ2+2 θ3
- Let X1,X2,…,Xn be a random sample of size n from N(μ,σ2) obtain (1-α)%
confidence interval for σ2 using the large sample behavior of MLE
- Find the Bayes Estimator of parameter p of a Binomial Distribution with X successes
out of n trials given that the prior distribution of p is a Beta distribution with
parameter α and β.
SECTION – C
Answer any two questions: (2×20=40Marks)
- i. Establish: If UMVUE exists for a parametric function , It must be essentially
unique.
- Obtain UMVUE of θ(1-θ) using a random sample of size n from B(1,θ).
- Let X1,X2,…,Xn be a random sample from N(µ,σ2). Find Cramer-Rao lower bound for
estimating a) µ b) σ2 c) µ+ σ d)
- Define Consistent Estimator and Establish the sufficient condition for Consistency.
- Establish: δ*Ug is QA-Optimal if and only if each component of δ* is UMVUE.
- Let X1,X2,…,Xn be a random sample from N(μ,σ2),μR, σ2>0. Obtain MLE of (μ,σ2)
- Explain Bootstrap and Jackknife Methods.