A Generalized Autoregressive Conditional Heteroskedasticity Examination of the Relationship between Trading Volume and Conditional Volatility in the Tunisian Stock Market: Evidence for the Information Flow Paradigm

This paper empirically examines the relationship between trading volume and conditional volatility of returns in the Tunisian stock market within the framework of the mixture of distribution hypothesis (MDH) and the sequential information arrival hypothesis (SIAH). Through this study, we especially aim to test the volatility persistence degree without volume, with contemporaneous volume, and with lagged volume. Our empirical analysis is based on daily data related to the 43 most active and dynamic securities traded from January 2, 2008 to June 29, 2012. Our daily analysis reveals several results. Firstly, we confirm the strong positive relationship between trading volume and returns conditional volatility issued from generalized autoregressive conditional heteroskedasticity (GARCH (1,1)) model. Secondly, according to the theoretical predictions of the MDH, we show that including contemporaneous trading volume in the conditional variance equation significantly reduces volatility persistence. Thirdly, through the addition of the lagged volume in the conditional variance equation, we show that volatility persistence remains in the whole at a high level and close to that obtained from the GARCH (1,1) model without trading volume, and also at a higher level than that resulting from the addition of the contemporaneous volume. Our results thus do not support the implications of the SIAH.