On empirical bayes rules for selecting the best hypergeometric population

In this paper, we consider k hypergeometric populations πi = π(Mi,mi,si), i=1,…,k, where Mi is the number of units in population πi, mi is the number of units selected from πi, and si is the number of defective units in πi . Let . A population πi with is considered as a best population. We are interested in selecting the best population and the best population compared with a control. It is assumed that the unknown parameteres si i = 1,…,k, …