Bayesian Inference on the Type-II Extreme Value Distribution Parameters

Abstract In this paper, Bayesian inference is applied for estimating the type-II extreme value distribution parameters and reliability when the experimental data are collected under type-II progressive censored samples. Via Monte Carlo simulations, the statistical properties of the estimations under the non-informative prior (NIP) are compared with those under the informative prior (IP), when prior information on the unreliability level at a fixed time is introduced. This study has shown that the use of prior information outperforms the statistical properties of the estimations, even when only a modest amount of prior information on the process parameters is available. Finally, a numerical example is given to illustrate the inferential method developed in this paper.