On the Bayesian Estimation for two Component Mixture of Maxwell Distribution, Assuming Type I Censored Data
Syed Mohsin, Ali Rashed Kazmi, Muhammad Aslam, Sajid Ali · 2012
Mixture models constitute a finite and infinite number of components that explain different datasets. However there are many situations where mixture models comprise an interesting sketch of different aspects. In this study we explore the idea of mixture density under Type I censoring scheme. We model a heterogeneous population by means of two components mixture of the Maxwell distribution. The parameters of the Maxwell mixture are estimated and compared using the Bayes estimates under the square error loss function and precautionary loss function. A censored mixture data is simulated by probabilistic mixing for the computational purpose. Closed form expressions for the Bayes estimators and posterior risk are derived for the censored sample as well as for the complete sample. Some interesting comparison and properties of the estimates are observed and presented. A real life data application has also been discussed.