Charting By Machines

Abstract

We test the efficient market hypothesis by using machine learning to forecast future stock returns from historical performance. These forecasts strongly predict the cross section of future stock returns. The predictive power holds in most subperiods, is strong among the largest 500 stocks, and is distinct from momentum and reversal. The forecasting function has important nonlinearities and interactions and is remarkably stable through time. Our research design ensures that our findings are not a result of data mining. These findings question the efficient market hypothesis and indicate that investment strategies based on technical analysis and charting may have merit.