Bayesian parameter estimation of beta log Weibull distribution under Type II progressive censoring

Abstract

In this article we discuss classical and Bayesian estimation procedures for estimating the unknown parameters as well as the reliability and hazard functions of the Beta Log Weibull distribution under Type II progressive censoring scheme. Conjugate and non-informative priors under squared error and entropy loss function are considered for Bayesian analysis. Complex and intractable integral functions are numerically approximated through Lindley approximation method and Markov Chain Monte Carlo techniques. Asymptotic confidence, Bootstrap confidence, Bayesian credible and highest posterior density intervals along with their coverage probability are also evaluated. A comprehensive simulation study to compare the performances of the maximum likelihood and Bayes estimates in terms of their mean squared errors is carried out. A real data set illustrates theoretical results.

Subject Classification:

• 62 F15

Keywords:

• Type II progressive censoring
• Bayes estimates
• Bootstrapping
• Credible intervals
• Markov Chain Monte Carlo technique
• Lindley Approximation method