Bayesian Estimation from Censored Data with Incomplete Information

Phillips and Sweeting (Journal of the Royal Statistical Society B 1996; 58:775–783) considered estimation of the parameter of the exponential distribution with censored failure time data when there is incomplete knowledge of the censoring times. It was shown that, under particular models for the censoring mechanism and censoring errors, it will usually be safe to ignore such errors provided they are not expected to be too large. The results depended on using an inverted Gamma model for the conditional distribution of the concomitant variable which provided the supplementary information used when the failure times were censored. The effect of alternative assumptions about knowledge of the censoring values on the estimation of failure rate is investigated through the use of Bayesian analysis. By putting prior distributions on the critical parameters of the censoring mechanism it is possible to incorporate and investigate the influence of the use of prior knowledge. Copyright © 2004 John Wiley & Sons, Ltd.