Does the Use of Downside Risk-Adjusted Measures Lead to Better Future Performance?

Abstract

This paper investigates how past performance, as measured by alternative risk-adjusted measures is informative of future performance. Performance results based on the Sharpe Ratio and several downside risk-adjusted performance measures (the Sortino ratio, excess return on Cornish-Fisher VaR, excess return on VaR, and excess return on Expected Shortfall) are compared. VaR and Expected Shortfall are computed using Filtered Historical Simulation (FHS), a hybrid methodology that combines the use of volatility models with bootstrapping techniques and is robust to the stylized facts of time series data on financial asset returns.

Recent empirical evidence is somewhat controversial. Some studies suggest that the use of different risk-adjusted performance measures (traditional vs downside risk-adjusted) does not seem to matter for investment fund decisions (Eling & Schuhmacher, 2007; Eling, 2008) while others suggest that it matters at least for some of the measures (Ornelas et al. 2010; Zakamouline, 2010). In most of the cases, the observed level of statistical association between the Sharpe Ratio and the downside risk-adjusted measures of performance is not as low as would be expected.

However, even if in empirical research statistical tests are not able to capture significant differences between rankings based on different measures, the economic significance of these differences may be relevant for investors and fund managers. In fact, it is of interest to assess to what extent the ex ante use of each of these measures can lead to better ex post performance results, regardless of the correlation between performance rankings and measures.

Our sample includes portfolios of Large Cap US funds over the period 1999 to 2010. In each month, portfolios of funds are created based on the rankings of the funds resulting from the different risk-adjusted performance measures. A estimation window of five years is used. The cumulative return of the top performing funds is then compared with the bottom performing funds.

Keywords: Downside risk-adjusted performance measures, Filtered historical simulation, fund performance predictability