Quantifying Backtest Overfitting in Alternative Beta Strategies

Abstract

We investigate the biases in the backtested performance of “alternative beta” strategies using a sample of 215 commercially promoted trading strategies across five asset classes. Our results lend support to the cautions in recent literature regarding backtest overfitting and lack of robustness in trading strategy performance during the ”live” period (out of sample). We report a median 73% deterioration in Sharpe ratios between backtested and live performance periods for the strategies in our sample. We establish a link between performance deterioration and strategy complexity, with the realized reduction in live vs. backtested Sharpe ratios of the most complex strategies exceeding those of the simplest ones by over 30 percentage points. The robustness of strategy exposure to risk factors varies between asset classes and strategies, and appears reasonable in equity volatility and FX carry strategies, but quite weak in the equity value strategy in particular.

Keywords: Alternative beta, Smart beta, Risk premia, Risk factor, Factor investing, Trading strategies, Index strategies, Investment strategies, Quantitative investment strategies, QIS, Sharpe ratio, Backtest, Overfitting, Data mining, Multiple test

JEL Classification: G11, G12, G14, G17, G23, G24