出版社：International Digital Organization for Scientific Information Publications
摘要：This article deals with the statistical inference for a step-stress partially accelerated life tests with twostress levels under progressive type-II censoring. The lifetime of the test units is assumed to followdistributions having power hazard function (DPHF). The maximum likelihood (ML), Bayes and parametricbootstrap methods are used for estimating unknown parameters of DPHF and the acceleration factor. Basedon normal approximation to the asymptotic distribution of MLEs, the approximate confidence intervals for theparameters and the acceleration factor are derived. In addition, two bootstrap confidence intervals are alsoproposed. The classical Bayes estimates cannot be obtained in explicit form, so we propose to apply theMarkov chain Monte Carlo (MCMC) method to tackle this problem, which allows us to construct the credibleinterval of the involved parameters. Finally, analysis of a simulated data set has also been presented to illustratethe proposed estimation methods.
关键词：Distributions having power hazard function Step-stress partially accelerated life test model;Bootstrap methods Bayesian estimation MCMC method
