Estimating certain functions of normal parameters
Abstract
A general problem of estimating the expected value of a function f of a normal random variable with unknown mean and unknown variance, is considered. Under certain conditions on f, asymptotic formulas for risk functions are developed. In addition to this, admissibility of a family of estimators of the normal distribution function at a point and of the normal density function, within the class of scale equivariant procedures, is demonstrated.
Degree
Ph.D.
Advisors
Rukhin, Purdue University.
Subject Area
Statistics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.