Asset allocation with time series momentum and reversal

To capture the well documented time series momentum and reversal in asset price, we develop a continuous-time asset price model, derive the optimal investment strategy theoretically, and test the strategy empirically. We show that, by combining market fundamentals and timing opportunity with respect to market trend and volatility, the optimal strategy based on time series momentum of moving averages over short-time horizons and reversal significantly outperforms, both in-sample and out-of-sample, the S&P 500 and pure strategies based on either time series momentum or reversal only. The results are robust for different time horizons, short-sale constraints, market states, investor sentiment, and market volatility.