Bayesian Estimation for 3-Component Mixture of Generalized Exponential Distribution

Abstract

This paper presents the Bayesian analysis of 3-component mixture of generalized exponential distribution. The Bayesian analysis and maximum likelihood estimation of five parameters have been performed by assuming type-I right censored data. Monte Carlo simulation has been adopted for the comparison of Bayes estimates, posterior risks, maximum likelihood estimates and maximum likelihood risks. Furthermore, the study assesses the performance of Bayes and maximum likelihood estimates by using different sample sizes, proportion of mixture components, censoring rates and loss functions. The Bayes estimates are examined by using non-informative Jeffreys and uniform prior under square error loss function, precautionary loss function and DeGroot loss function. The maximum likelihood risks are obtained by using the Fisher information matrix.