A0475
Title: Testing the order of multivariate normal mixture models
Authors: Hiroyuki Kasahara - University of British Columbia (Canada)
Katsumi Shimotsu - University of Tokyo (Japan) [presenting]
Abstract: Finite mixtures of multivariate normal distributions have been widely used in empirical applications in diverse fields such as statistical genetics and statistical finance. Testing the number of components in multivariate normal mixture models is a long-standing challenge even in the most important case of testing homogeneity. A likelihood-based test is developed for the null hypothesis of $M$ components against the alternative hypothesis of $M+1$ components for a general $M\ge 1$. We derive the asymptotic distribution of the proposed EM test statistic under the null hypothesis and local alternatives and show the validity of the parametric bootstrap. The simulations show that the proposed test has a good finite sample size and power properties.
