This paper studies the association and the model construction of the Shenzhen and the Hong Kong stock markets. The data period is from January 4, 1999 to December 30, 2005. The empirical analyses indicate that there is a weaker association between the Shenzhen and the Hong Kong stock markets. We use a bivariate GARCH (1, 2) model with a dynamic conditional correlation (DCC) to evaluate the association and find that there exists an asymmetrical effect for the two stock markets. The result of the empirical analyses also shows that the Hong Kong stock market returns positively affect the Shenzhen stock market returns. And the volatility of the Hong Kong and the Shenzhen stock market returns interact with one another. The average of DCC coefficient of two stock market returns equals to t ρ =0.1315, and the coefficient of t ρ is significant under the 1% significance level. Besides, the Hong Kong stock market has an asymmetrical effect, but Shenzhen stock market has not asymmetrical effect during the research sample period. The explanatory ability of the bivariate GARCH(1, 2) model with a DCC is better than the model of the bivariate GARCH model with a constant conditional correlation. This result is consistent to the paper of Engle (2002)