Many data that exhibit asymmetrical behavior can be well modeled with skew-normal random errors.Moreover,data that arise from a heterogeneous population can be efficiently analyzed by a finite mixture of regression mo...Many data that exhibit asymmetrical behavior can be well modeled with skew-normal random errors.Moreover,data that arise from a heterogeneous population can be efficiently analyzed by a finite mixture of regression models.These observations motivate us to propose a novel finite mixture ofmode regression model based on amixture of the skew-normal distributions to explore asymmetrical data from several subpopulations.Thanks to the stochastic representation of the skew-normal distribution,we construct a Bayesian hierarchical modeling framework and then develop an efficient Markov chain Monte Carlo sampling algorithm to generate posterior samples for obtaining the Bayesian estimates of the unknown parameters and their corresponding standard errors.Simulation studies and a real-data example are presented to illustrate the performance of the proposed Bayesian methodology.展开更多
基金supported by the Academics and Technology Foundation of Kunming University of Science and Technology(Grants 2020YB208)supported by the Natural Science Research Foundation of Kunming University of Science and Technology(Grants KKSY201907003)supported by the National Natural Science Foundation of China(Grants 11861041).
文摘Many data that exhibit asymmetrical behavior can be well modeled with skew-normal random errors.Moreover,data that arise from a heterogeneous population can be efficiently analyzed by a finite mixture of regression models.These observations motivate us to propose a novel finite mixture ofmode regression model based on amixture of the skew-normal distributions to explore asymmetrical data from several subpopulations.Thanks to the stochastic representation of the skew-normal distribution,we construct a Bayesian hierarchical modeling framework and then develop an efficient Markov chain Monte Carlo sampling algorithm to generate posterior samples for obtaining the Bayesian estimates of the unknown parameters and their corresponding standard errors.Simulation studies and a real-data example are presented to illustrate the performance of the proposed Bayesian methodology.
