BAYESIAN VARIABLE SAMPLING PLANS FOR THE EXPONENTIAL-DISTRIBUTION WITH UNIFORMLY DISTRIBUTED RANDOM CENSORING

Lam (The Statistician (1990) 53-67; Ann. Statist., (1994) 696-711) studied the models of single variable sampling plan under Type II and Type I censoring. In this article, a similar model for the case of random censoring is considered. The lifetime of an item in a large batch is assumed to be exponentially distributed and the censoring time is uniformly distributed over a known interval. Assume further that the loss function is a polynomial function. An explicit expression for the Bayes risk is derived. Moreover, a finite algorithm and a discretization method are suggested to determine an optimal sampling plan approximately.