Estimation and prediction for Type-I hybrid censored data from generalized Lindley distribution

Abstract

This paper consider the problems of estimation and prediction using Type-I hybrid censored lifetime data that follow generalized Lindley distribution. Maximum likelihood estimators as well as Bayes estimators have been proposed for estimating the parameters and reliability characteristics from the generalized Lindley distribution. Since posteriors are not in closed forms, Markov Chain Monte Carlo techniques such as Gibbs sampler and Metropolis-Hastings algorithm have been utilized to explore the properties of the posteriors. Monte Carlo simulation study has been carried out to compare the classical and Bayesian estimation methods. One and two sample predictive posteriors of future order statistics are also derived on the basis of Type-I hybrid censored data. Finally, a set of real data is analysed for illustration.