An application of the multivariate skewnormal distribution for Dirichlet processes in mixed-effects modeling

An application of the multivariate skewnormal distribution for Dirichlet processes in mixed-effects modeling

In this paper, we consider a semi-parametric Bayesian model for analyzing of panel data. This is done by introducing a skew Dirichlet process (SDP) as a generalization of the usual Dirichlet process (DP). The SDP is specified by introducing a multivariate skew-normal distribution as a base for the DP instead of the normal distribution. The proposed approach which is applied for the random-effects, can efficiently deal with modeling issues in a wide range of non-normally distributed random-effects. We adopt Gibbs sampling techniques to achieve the parameter estimate real data set is analyzed and tested by several hypothetical models to illustrate the usefulness of the proposed approach