On Empirical Bayes Testing in a Discrete Life Model with Random Censoring

Abstract

This article studies an empirical Bayes test in a discrete life model based on random censored data. A monotone empirical Bayes test is constructed and its associated asymptotic optimality is also investigated. It is shown that is asymptotically optimal, and the regret of converges to zero at a rate for some positive values c ₁ and c ₂, where n is the number of past data available when the present testing problem is considered.

Keywords:

• Asymptotic optimality
• Random censoring
• Rate of convergence
• Regret
• Two-action problem

Mathematics Subject Classification:

• 62C12