From Moments, Co-Moments and Mean-Variance weights to Copula Portfolio Allocation

Riccetti (2010) finds that the use of copulas can be useful in the macro asset allocation process, to increase investors' utility. Thus, easy methods to reconstruct the portfolio allocation obtained with a GARCH-copula model without estimating the copula, could be important for an asset manager. In this paper I analyse two models: the first reconstructs the portfolio weights obtained with the copula model using moments and co-moments of the returns, while the second adds to moments and co-moments also the portfolio weights obtained with the mean-variance approach. The in-sample and the out-of-sample analyses show that it is possible to have an approximation of the weights obtained by a copula model using moments and co-moments of returns: the first appears easier (because the weights of the Markowitz model are not needed), while the second is more accurate.