Filter rules based on price and volume in individual security overreaction

I present evidence of predictability in a sample constructedto minimize concerns about time-varying risk premia and market-microstructureeffects. I use filter rules on lagged return and lagged volumeinformation to uncover weekly over-reaction profits on large-capitalizationNYSE and AMEX securities. I find that decreasing-volume stocksexperience greater reversals. Increasing-volume stocks exhibitweaker reversals and positive autocorrelation. A real-time simulationof the filter strategies suggests that an investor who pursuesthe filter strategy with relatively low transaction costs willstrongly outperform an investor who follows a buy-and-hold strategy.