Approximation of Asymmetric Multivariate Return Distributions

We develop a new method to approximate the asymmetric multivariate probability density function (pdf) of financial asset returns by using series expansions; a rate of convergence for the mean absolute error of this approximation is also provided. We then propose the method of maximum likelihood and the generalized method of moments to estimate the parameters of the approximated pdf. A Monte-Carlo experiment corroborates the feasibility of our approach.