Stock Market Volatility and Trading Volume : An Emerging Market Experience

This study investigates the relationship between daily returns and trading volume for 30 stocks included in the Istanbul Stock Exchange National-30 index. Study findings reveal that GARCH model is an appropriate model to mimic the conditionality of the second moments. The study finds that the persistency of conditional volatility is high and very close to unity, implying that current information can be used to predict future volatility. It also finds no leverage effect contradicting other studies reporting leverage effect. Speculative bubbles resulting from economic factors not justifying the rise in the market may cause these results. When trading volume is included in the analysis, the study finds that even though the persistence of the conditional volatility is present, it is lower with the introduction of volume. Finally, the decomposition of volume into expected and unexpected components shows that the expected component of volume significantly explains the variation in volatility. A vast majority of firms exhibit a decline in their volatility persistence. This result suggests that surprises in trading volume do not convey all the information associated with trading volume and that most information release is predictable.