Bayesian inference from the Kumaraswamy-Weibull distribution based on censored samples with applications real data

Abstract

This paper is concerned with the Bayesian analysis for the Kumaraswamy-Weibull (Kum-W) distribution under type II censored samples. Approximate Bayes estimates are computed using the Gibbs sampling procedure. This procedure generates samples from the posterior distributions. The approximate Bayes estimators are obtained under the assumptions of non-informative priors. Also, using Bayesian framework, the posterior density function, the predictive density for a single future response, i th ordered future response, and several future responses are derived under type II doubly censored samples. The predictive means, standard deviations, prediction intervals, and the shape characteristics for a single future response are determined. Applications to real data sets are utilized to illustrate the potentiality of the Bayesian analysis and the predictive results.