Abstract
In this paper, new classes of rules are proposed for selecting from a given set of populations, a subset containing the best population. The rules have the following property: The probability that the selected set contains the best population is greater than or equal to P*, a preassigned number. The derivation of the rules is based on Schur concave functions. The rules apply to various types of distributions such as normal and gamma.
Get full access to this article
View all access options for this article.