Sequentual First-Crossing Look-Ahead Procedure for Selecting a Population with the Largest Meanin Normal-Normal Model

Abstract

The problem of statistical selection of a population with the largest mean value is considered. We introduce a sequential selection procedure, which we call first-crossing look-ahead (FCLA), for a normal-normal Bayesian setting of the problem, where variances of the populations are supposed to be the same and known, and the means are realizations of prior normal random variables with known distribution parameters. The paper includes the definition of the procedure with some basic analytical results, the results of numerical simulations, and a numerical performance comparison (in terms of sample size) with one of known efficient selection procedure for an indifference-zone setting of the selection problem.