The predictive power of "Head-and-Shoulders" price patterns in the US stock market

We use the pattern recognition algorithm of Lo, Mamaysky, and Wang (2000) with some modifications to determine whether "head-and-shoulders" (HS) price patterns have predictive power for future stock returns. The modifications include the use of filters based on typical price patterns identified by a technical analyst. With data from the S&P 500 and the Russell 2000 over the period 1990-1999 we find little or no support for the profitability of a stand-alone trading strategy. But we do find strong evidence that the pattern had power to predict excess returns. Risk-adjusted excess returns to a trading strategy conditioned on "head-and-shoulders" price patterns are 5-70/o per year. Combining the strategy with the market portfolio produces a significant increase in excess return for a fixed level of risk exposure.