沪深股市动态溢出效应与动态相关性的实证研究--基于长记忆VAR-BEKK(DCC)-MVGARCH(1，1)模型

Empirical Research on Dynamic Spillover Effect and Correlation in Shanghai and Shenzhen Stock Markets:Based on Long Memory VAR-BEKK(DCC)-MVGARCH Model

利用长记忆VAR-BEKK-MVGARCH(1,1)模型和VAR-DCC(1,1)-MVGARCH(1,1)模型,对沪深股市的动态均值溢出效应、动态波动溢出效应和动态相关性进行了实证检验。结果表明:沪深股市收益率的均值溢出效应和波动溢出效应,在2000年之前基本不存在,仅在2000年之后才比较显著,且表现为双向的传导关系。沪深股市收益率间表现出一定程度的动态相关性,除股市成立初期表现出负相关外,总体呈现出正相关,且相关性逐步提高,近年来稳定在0.8到0.9之间。

Dynamic mean spillover effect, volatility spillover effect and correlation of Shanghai and Shenzhen stock markets returns were empirically verified by using long memory VAR-BEKK-MVGARCH（1,1） and VAR-DCC （1,1）-MVGARCH （1,1） models. As results have shown, before 2000, mean and volatility of Shanghai and Shenzhen stock market were relatively independent. After that, there are double direction relations of mean and volatility spillover effect between Shanghai and Shenzhen stock market returns. Additionally,empirical results suggest that dynamic correlation exists in Shanghai and Shenzhen stock market returns to a certain extent, that the correlation tends to be positive except in the period of stock market founding and that it increased gradually, and changed stably between 0. 8 to 0. 9.