Operating Characteristics of Subset Selection Rules for Exponential Population Threshold Parameters

Abstract

This article provides the operating characteristics (OCs) of two subset selection rules for exponential populations having a common known scale parameter and possibly differing threshold parameters. One selection rule is based on the minimum sample values and the other is based on the mean (or sum) of the sample values. The random samples drawn from the populations are of equal size. The goal of the selection rules is to choose a subset of the populations such that the population possessing the largest threshold parameter (the “best” population) is contained in the subset with a probability no less than a user prescribed value P *. A correct selection occurs if the best population is contained in the selected subset. The OCs are the probability of a CS and the expected size of the selected subset. The OCs are calculated and compared for several formulations of the selection rules and for two threshold parameter configurations—slippage, and equi-spaced. The computer R-codes for all calculations are given in the Appendices.