Testing Homogeneity of Mixture of Skew-normal Distributions Via Markov Chain Monte Carlo Simulation

摘要

The main purpose of this study is to intoduce an optimal penalty function for testing homogeneity of finite mixture of skew-normal distribution based on Markov Chain Monte Carlo (MCMC) simulation. In the present study the penalty function is considered as a parametric function in term of parameter of mixture models and a Baysian approach is employed to estimating the parameters of model. In order to examine the efficiency of the present study in comparison with the previous approaches, some simulation studies are presented.