Stock return predictability: Evidence from moving averages of trading volume

This study investigates the role of moving averages of trading volume on asset pricing. We find that the distance between short- and long-term moving averages of trading volume (MAVD) strongly and negatively predicts the cross-section of stock returns in the Chinese stock market. This predictive power is robust after controlling for other firm characteristics, well-known risk factors and market timing, and goes well beyond the price-based distance predictor. Moreover, the MAVD effect diminishes as portfolio holding months move further away from the portfolio formation month and even reverses at the end of the second year, which suggests that stock market overreacts to the information from MAVD and the resulting mispricing is gradually corrected. Our results also show that the MAVD effect is stronger in stocks with high limits of arbitrage and more investor attention, as well as in the periods of high sentiment and high investor overconfidence, which is consistent with behavioral mispricing explanations. Furthermore, we find that the MAVD effect is likely to be attributable to individual speculative trading behavior. Finally, the evidence indicates that the predictive power of MAVD is more pronounced among high volatility stocks rather than among low volatility stocks.