Hidden Risks in Mean-Variance Optimization: An Intergrated-Risk Asset Allocation Proposal

Abstract

The traditional mean-variance asset allocation approach considers the volatility of returns as the only risk factor. However, some asset classes present other types of risk, and so offer a premium embedded in their returns to compensate for the additional risk. If those risk premia are not taken into account in the analysis, the results of the model tend to be distorted, with for instance credit and more negative skewed portfolios dominating the results. This paper has two main objectives. The first is to propose a novel performance measure that takes into account both skewness and credit risk, that we call Adjusted for Skewness and Credit Ratio (ASCR). Second, we evaluate the performance and hidden risks for a conservative long-term investor, considering the effects of the inclusion of hedge funds, corporate and high yield bonds in a meanvariance optimization for strategic asset allocation. Our results verify that a simplistic mean variance asset allocation is distorted in the presence of credit and negative skewed assets. Moreover, the proposed Adjusted for Skewness and Credit Ratio shows that while the effect of skewness is weak, credit risk has a very strong impact in the asset allocation solution.

Keywords: Integrated Risk Measure, Skewness, Credit Risk

JEL Classification: G11, G15