Trading volume and serial correlation in stock returns: a threshold regression approach

We extend the analysis of Campbell et al. (1993) on the relationship between the first-order daily stock return autocorrelation and stock market trading volume by allowing abrupt and smooth transition structures using lagged stock returns as a transition variable. Using U.S. stock market data, we find the evidence supporting the nonlinear relationship characterized by a stronger return reversal effect on a high-volume day combined with low lagged stock returns.