Empirical bayes testing for uniform distributions with random censoring

In this article, we study empirical Bayes testing for uniform distributions based on randomly censored data. We construct an empirical Bayes test δn, and study its associated asymptotic optimality. It is shown that under some conditions on the prior distribution π, δn is asymptotically optimal and its associated regret converges to zero at a rate O(n−r/(r+1)), where r is a positive integer, depending on a condition on π and n is the number of past data available when the present testing problem is considered. © 2008 Taylor & Francis Group, LLC. All rights reserved.