E-Bayesian Estimation for a Constant-Stress Partially Accelerated Life Test Based on Burr-X Type-I Hybrid Censored Data

This paper is concerned with a constant-stress partially accelerated life test (CSPALT) model whose items are allocated at either normal use condition or under accelerated use condition. The lifetimes of items are assumed to follow Burr-X distribution based on Type-I hybrid censoring scheme. Two methods of estimation namely, E-Bayesian and Bayesian estimation methods are used for estimating the distribution parameter and the acceleration factor. These estimates are derived by using LINEX and squared error loss (SEL) functions. To obtain estimates of Bayesian and E-Bayesian methods, Markov chain Monte Carlo (MCMC) method is used. For illustration, a real-life data set is presented and applied to the CSPALT model. In the end, a comparison between proposed methods is carried out to investigate the performance of methods.