Portfolio Allocation using Multivariate Variance Gamma

In this paper we investigate empirically the effect of using higher moments in portfolio allocation when parametric and non parametric models are used. The non parametric model considered, in this paper, is the sample approach while the parametric one is constructed assuming Multivariate Variance Gamma (MVG henceforth) joint distribution for assets returns. We consider the MVG models proposed by Madan and Seneta (1990), Semeraro (2006) and Wang (2009). We perform an out-of-sample analysis comparing the optimal portfolios obtained using the MVG models and the sample approach. Our portfolio is composed of 18 assets selected from the S&P500 Index and the dataset consists in daily returns observed in the period ranging from 01/04/2000 to 01/09/2011.