Information-Based Trading and Autocorrelation in Individual Stock Returns

Applying a recently developed approach, the paper estimates the daily arrival rates of buy and sell orders originated from different trading motives for each stock in a sample of NYSE-listed companies. Based on these arrival rates, it shows that stock return tends to continue on consecutive days when privately-informed trading prevails, leading to positive return autocorrelation. But return is more likely to reverse itself on days with continuous trading on dispersion in beliefs, leading return autocorrelation to be more negative. Contrarian trading strategies conditional on daily measures of investment disagreement can yield economically and statistically significant excess returns