Return Predictability and Market-Timing: A One-Month Model

Vol. 17, No. 3, 2019
Blair Hull, Xiao Qiao and Petra Bakosova

We use weighted least squares to combine 15 diverse variables to build a predictive model for the one- month-ahead market excess returns.We transform our forecasts into investable positions to form a market-timing strategy. From 2003 to 2017, our strategy had 16.6% annual returns with 0.92 Sharpe ratio and 20.3% maximum drawdown. In comparison, the S&P 500 had annual returns of 10%, 0.46 Sharpe ratio, and maximum drawdown of 55.2%.We also combine our one-month model with the six-month model of Hull and Qiao (2017). The combined model had 15% annual returns, Sharpe ratio of 1.12, and maximum drawdown of 14%