Author : Faisal Hathot, Samira
Bayesian Fixed Sample Size Procedures for Selecting the Better of Two Exponential Populations With General Loss Function
journal of kerbala university,
Volume 3, Issue 3, Pages 78-86
In this paper an optimal (Bayesian) fixed sample size procedure for selecting the better of two Exponential populations is proposed and studied . Bayesian decision-theoretic approach with general loss function and Gamma priors are used to construct this procedure . A suboptimal procedure that is based on posterior estimate of the parameters and a method of obtaining an approximation to the optimal procedure using stirling’s formula are also presented. Comparisons among these procedures are made using performance characteristics such as Bayes risk .