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

INTEREST RATE MODELS, ASSET ALLOCATION AND QUANTITATIVE TECHNIQUES FOR CENTRAL BANKS AND SOVEREIGN WEALTH FUNDS, Arjan Bastiaan Berkelaar, Joachim Coche and Ken Nyholm, eds.

23 Pages Posted: 26 Nov 2008 Last revised: 2 Oct 2010

Jose L. B. Fernandes

Universidade de Brasília

Jose Renato Haas Ornelas

Banco Central do Brasil

Date Written: October 28, 2008

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

Suggested Citation

Fernandes, Jose L. B. and Ornelas, Jose Renato Haas, Hidden Risks in Mean-Variance Optimization: An Intergrated-Risk Asset Allocation Proposal (October 28, 2008). INTEREST RATE MODELS, ASSET ALLOCATION AND QUANTITATIVE TECHNIQUES FOR CENTRAL BANKS AND SOVEREIGN WEALTH FUNDS, Arjan Bastiaan Berkelaar, Joachim Coche and Ken Nyholm, eds. . Available at SSRN: https://ssrn.com/abstract=1307087

Jose Luiz Barros Fernandes

Universidade de Brasília ( email )

Campus Universitário Darcy Ribeiro
Brasília, Distrito Federal 70910-900
Brazil

Jose Renato Haas Ornelas (Contact Author)

Banco Central do Brasil ( email )

P.O. Box 08670
SBS Quadra 3 Bloco B - Edificio-Sede
Brasilia, Distrito Federal 70074-900
Brazil

HOME PAGE: http://www.bcb.gov.br

Paper statistics

Downloads
305
Rank
80,325
Abstract Views
2,132