Center for Risk Management Research

(from the introduction...)

In this paper, we combine the innovations described above in an empirical study of expected shortfall optimization with Factor-Based Extreme Risk. We avoid the issue of forecasting mean return by comparing minimum expected shortfall to minimum variance portfolios. Our study is carried out for the US, UK, and Japanese equity markets and it uses Barra Style Factors (Value, Growth, Momentum, etc.). We show that minimizing expected shortfall generally improves performance over minimizing variance, especially during down-markets, over the period 1985-2010. The outperformance of expected shortfall is due to intuitive tilts towards protective factors like Value, and away from aggressive factorslike Growth and Momentum. The outperformance is largest for the expected shortfall at relatively low confidence levels, which measures distributional asymmetry rather than the extreme losses.