<p>This paper introduces a statistical analysis of the progressive-stress accelerated life test (PSALT) for the Perk distribution under adaptive progressive type-II hybrid censoring (AP-II-HC). The cumulative exposure model is used as a progressive-stress model. Many estimation methods are used to estimate the distribution’s parameters, including classical and Bayesian methods. We use the Metropolis Hasting algorithm (Metropolis et al. in J Chem Phys 21:1087–1092, 1953) to generate samples because the posterior is not from a well-known distribution. The unknown parameters’ asymptotic and bootstrap confidence intervals (CIs) are estimated. Furthermore, the reliability function of the distribution is estimated. A real data set is analyzed to clarify the methods proposed in this paper. A practical application is applied as a lifetime example to clarify the importance of the work. Finally, some interesting conclusions are drawn.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Statistical inference on progressive-stress accelerated life testing for the Perk distribution under adaptive type-II hybrid censoring scheme

  • Eslam Hussam,
  • Ehab M. ALMetwally

摘要

This paper introduces a statistical analysis of the progressive-stress accelerated life test (PSALT) for the Perk distribution under adaptive progressive type-II hybrid censoring (AP-II-HC). The cumulative exposure model is used as a progressive-stress model. Many estimation methods are used to estimate the distribution’s parameters, including classical and Bayesian methods. We use the Metropolis Hasting algorithm (Metropolis et al. in J Chem Phys 21:1087–1092, 1953) to generate samples because the posterior is not from a well-known distribution. The unknown parameters’ asymptotic and bootstrap confidence intervals (CIs) are estimated. Furthermore, the reliability function of the distribution is estimated. A real data set is analyzed to clarify the methods proposed in this paper. A practical application is applied as a lifetime example to clarify the importance of the work. Finally, some interesting conclusions are drawn.