Empirical Bayes Sequential Estimation for finite population
by
Ejaz Ahmed
University of Windsor, Ontario, Canada
Coauthors: Mohannad Al-Khasawneh

We consider the Bayes and empirical Bayes estima- tion of the current population mean of a finite population when the sample data from other similar (m-1) finite populations are available. We investigate a general sequential sampling procedure in which the observations from the current population are taken one at the time until a stopping rule is satisfied. We find the optimal stopping rule which generates optimal linear Bayes estimator of the current pop- ulation mean under a squared error loss plus a sampling cost. The corresponding empirical Bayes estimates are obtained by replacing the unknown hyperparameters by their respective consistent estimates. A Monte Carlo study is conducted to evaluate the performance of the proposed empirical Bayes procedure as compared to the usual Bayes procdure.

Date received: February 18, 2007