Trading Volume and Price Variability : Evidence on Lead-Lag Relations from Granger-Causality Tests

We test for Granger-causality between trading volume and price volatility. We modify the standard regression procedure in several ways. We take the first difference of the logarithmic transformation of each series to account for potential nonstationarity in the data. We isolate the time series structure of each series and test for causality using pre-whitened residuals to eliminate problems associated with autocorrelation in the data. Finally we test for causality in both mean and variance to account for the presence of time varying variance (ARCH) in both series. Our results provide strong evidence that price changes lead (cause) volume in the Granger-causality sense. There is no evidence that volume causes volatility