In this paper we study the dynamic relationship between trading volume, volatility, and stock returns at international stock markets. We test a number of theoretical models which suggest that the trading volume and volatility can predict future behavior of stock returns. Our analysis uses both semi-nonparametric (Flexible Fourier Form) and parametric techniques. Our findings suggest that the main factor driving the magnitude of the return reversals is stock market volatility and not trading volume. First, apart from a direct effect on expected returns with mixed signs, we find no evidence of the trading volume affecting the serial correlation of stock market returns, as predicted by Campbell et al. (1993) and Wang (1994). Second, the stock market volatility has a negative and statistically significant impact on the serial correlation of the stock market returns, consistent with the “positive feedback” trading model of Sentana and Wadhwani (1992). Third, the lagged trading volume is positively related to the stock market volatility, supporting the “information flow” theory (Clark, 1973). Moreover, we find that taking into account both trading volume and volatility improves the accuracy of the out-of-sample forecasts of the stock market behavior.
Available at: http://works.bepress.com/leon_zolotoy/2/