Constant conditional correlation in a bivariate GARCH model: evidence fromthe stock markets of China

In this paper we examine the behaviour of stock returns in two emerging mar kets of China. These are the Shanghai and Shenzhen markets. It is found tha t both markets suffer from negative mean returns on Monday and Tuesday, but positive returns on Friday. In addition, we employ the bivariate GARCH mod el of Bollerslev [T. Bollerslev, Review of Economics and Statistics 72 (199 0) 498-505] to capture the co-movements of stock returns between the market s. However, the information matrix test statistic does not support the null hypothesis of a constant conditional correlation in the stock returns. (C) 1999 IMACS/Elsevier Science B.V. All rights reserved.