Bayesian Estimators of the Parameters and Reliability Function from Mixed Exponentially Distributed Time-Censored Life Test Data

Bayes estimators of the parameters and the reliability function are obtained for the Mendenhall and Hader (1958) model where the population is made up of a mixture of two exponentially distributed subpopulations and the test is terminated after a predetermined time T hours. Based on a Monte-Carlo study these estimators are compared with their maximum likelihood counterparts. Appropriate Pearson-type curves are fitted to the sampling distributions of Baves estimates and a ‘goodness of fit’ test performed.