Bayesian Analysis of Power Generalized Weibull Distribution

The present paper focuses on Bayesian estimation of the unknown parametric functions for power generalized Weibull distribution, using Lindley and Markov chain Monte Carlo approximations, under Type II censoring. Bayes estimates of the unknown parametric functions are derived under natural conjugate prior and non-informative prior set up assuming squared error and linear exponential losses respectively. Comparative simulation studies of Bayes estimators with the corresponding maximum likelihood estimates are undertaken for efficiency assessment. The issue of model fit is addressed through a classical real data set. © 2018, Springer Nature India Private Limited.