Copulas are popular as models for multivariate dependence because they allow the marginal densities and the joint dependence to be modeled separately. However, they usually require that the transformation from uniform marginals to the marginals of the joint dependence structure is known. This can only be done for a restricted set of copulas, e.g. a normal copula. Our work introduces generalized copulas as flexible multivariate density estimators which also allow the marginal densities to be modeled separately from the joint dependence, as in all copula estimators, but overcomes the lack of flexibility of most popular copula estimators. An iterative scheme is proposed for estimating generalized copulas and its usefulness is demonstrated through examples. We revisit the mixture of normals (factor analyzers) and mixture of t (factor analyzers) models, and develop efficient variational Bayes algorithms for fitting them in which model selection is performed automatically. Based on these mixture models, we construct four instances of generalized copulas which are far more flexible than current popular copula densities, and outperform them in several real data sets. We also introduce a new criterion for selecting the best univariate density estimator among a class of competitors, which is important when fitting the marginals. This talk is based on a joint work with P. Giordani, R. Kohn and X. Mun.


Dr Minh-Ngoc Tran

Research Area

Statistics Seminar


Australian School of Business, UNSW


Fri, 27/04/2012 - 4:00pm to 5:00pm


OMB-145, Old Main Building, UNSW Kensington Campus