Abstract: The integration of market risk and liquidity risk facilitates the investors' comprehensive management of risks when theytrade their stocks. Considering the time variation, heteroscedasticity and tail characters of market risk and liquidity risk, GARCHEVTmethod is used for the modeling of these properties. Three types of Archimedean Copula are used to investigate the dependencestructure between two kinds of risks. The results indicate that the two kinds of risks show stronger dependence in the tails and thetwo tails are symmetric. The measurement based on such dependence structure performs better than traditional VaR and the VaRwithout regard to the dependence structure.