ABSTRACT
Work on empirical Bayes problems with randomly censored data appeared in Susarla and van Ryzin (1978). They treated the empirical Bayes decision problems using a Dirichlet process prior with an unknown parameter α. Susarla and van Ryzin (1986) also investigated the empirical Bayes approach to some decision problems with randomly censored data obtained from an exponential family. However, they only studied the asymptotic optimality of the empirical Bayes rules. The associated rates of convergence were not investigated. Recently, Liang (2004, 2006) has studied some empirical Bayes rules for exponential distributions based on randomly censored data. He assumed that the random censoring scheme follows a fully known exponential distribution.
