Name: David (Jianjun) Li

Firstname: David (Jianjun)

Title: Biometrician

Institution: Merck Research Labs

Street: 785 Jolly Road

City: Blue Bell

Zip-Code: 19422

Country: USA

Phone: 484-344-2101

Fax: 484-344-7105


Authors: Shanti S. Gupta and David (Jianjun) Li

Title: On empirical Bayes procedures for selecting good populations in positive exponential family

Abstract: The problem of selecting good ones compared with a control from $k (\ge 2)$ positive exponential family populations is considered in this paper. A nonparametric empirical Bayes approach is used to construct the selection procedures. It has been shown that the risks of the empirical Bayes procedures converge to the (minimum) Bayes risk with a rate of $O({1/n})$, where $n$ is the number of accumulated past observations at hand. Simulations were carried out to study the performance of the procedures for small to moderate values of $n$. The results of this study are provided in the paper.