Section: New Results

Mixture model for mixed kind of data

Participants : Christophe Biernacki, Matthieu Marbac-Lourdelle, Vincent Vandewalle.

A mixture model of Gaussian copula allows to cluster mixed kind of data. Each component is composed by classical margins while the conditional dependencies between the variables is modeled by a Gaussian copula. The parameter estimation is performed by a Gibbs sampler. An article has been presented to an international conference [48] and has been also submitted to an international journal [50] . Furthermore, an R package (MixCluster) is available on Rforge (see 5.12 ).