Intraday data and volatility models : evidence from Chinese stocks

This paper investigates the empirical relationship between intraday volatility and trading volume. Our primary dataset consists of 5-minute returns and trading volumes for the period between January 1, 2000 and December 31, 2002, for a subset of thirty-nine stocks from the Shanghai Stock Exchange 180 Index. Taking in consideration the excess kurtosis in high-frequency data, this study, unlike previous studies, estimated both the GARCH and the EGARCH model with generalized error distribution (GED) residuals. Similar to Rahman et al (2002), our results indicated that the GARCH (1,1) model best describes the volatility of intraday returns. Our results also show that the persistence in volatility remains in the intraday return series even after the lagged log-volume is included in the models as an explanatory variable. Our results further suggest volume as an information variable has quite a limited effect on the volatility of intraday returns in the Shanghai stock market. With respect to the asymmetric reaction of the predicted volatility to good and bad news, we find there is an adverse asymmetric reaction with good news increasing the volatility more than bad news, which is consistent with Giot (1999).