GAZI UNIVERSITY JOURNAL OF SCIENCE, cilt.29, sa.2, ss.335-341, 2016 (ESCI)
Finite mixtures of multivariate t distributions (Peel and McLachlan (2000)) were introduced as an alternative to the finite mixtures of multivariate normal distributions to model data sets with heavy tails. In this study, we define the finite mixtures of matrix variate t distributions as an extension of finite mixtures of multivariate t distributions. Mixtures of matrix variate t distributions can provide an alternative robust model to the mixtures of matrix variate normal distributions (Viroli (2011)) for modeling matrix variate data sets with heavy tails. We give an Expectation Maximization (EM) algorithm to find the maximum likelihood (ML) estimators for the parameters of interest. We also provide a small simulation study to illustrate the performance of the proposed EM algorithm for finding estimates.