A Case for Tail-Risk-Based Sharpe Ratios

Abstract

Surprisingly to many investors, low volatility tends to be accompanied with an undesirable risk characteristic: lower or negative skewness. A stock or fund can rank well based on the standard Sharpe ratio but low on enhanced tail-risk-based Sharpe ratios that account for non-normal returns, and vice versa. The authors quantify these economically meaningful ranking differences and show that skewness dominates other variables in explaining the ranking variations for the Conditional Value-at-Risk (CVaR)-based Sharpe ratio. Both skewness and serial correlation play important roles in the ranking variations for the maximum drawdown-based Sharpe ratio.