摘要：Over the last few decades, copulas have consistently gained significance in finance research, due to their usefulness in risk modeling. However, the idea of implicitly representing dependencies between multiple assets in a single mathematical entity is extremely useful in portfolio allocation models as well. While Church (2012) and many others have exploited these benefits, the efficiency of such frameworks in capturing the most essential features of financial data can still be enhanced. An obvious improvement would be to incorporate the fact that financial returns are generally asymmetric and skewed in nature, and therefore asymmetric (or skewed) margins can be used to describe them in a suitable copula framework. In this paper, we consolidate this idea with a GARCH(1,1) pre-whitening method that takes into account inter-temporal dependencies of returns, and use a utility maximization approach to find optimal portfolio allocation schemes. We show that the gains of optimal weighting, in terms of certainty equivalent returns, can be substantial for utility functions with reasonable risk aversion.
