22nd International Academic Conference, Lisbon

GRANGER CAUSALITY BETWEEN STOCK PRICES AND TRADING VOLUME: EVIDENCE FROM TURKEY

ELIF AKBEN-SELCUK

Abstract:

The objective of this study is to investigate the dynamic relation between daily BIST-100 index returns and percentage changes in Borsa Istanbul trading volume. A vector autoregression (VAR) model is constructed to test for Granger causality between stock prices and volume. The causality structure of the two variables is analyzed using the approach by Engle-Granger (1987). Analysis results show the existence of univariate causality from stock returns to changes in trading volume. This implies that past stock prices can be used to predict futures changes in trading volume. Furthermore, the results suggest that Borsa Istanbul is still inefficient since information contained in past prices is useful for making forecasts.

Keywords: Granger causality, stock prices, trading volume, Turkey.