中文摘要：

采用二次汇改后沪深300指数和人民币/美元汇率的日数据,结合AR-GARCH模型和极值理论POT模型,度量两个市场的95%和97.5%置信水平的Va R,并利用基于交叉相关函数的风险溢出检验方法,分析沪深300指数与汇率间的风险信息溢出。实证结果表明：在95%置信水平下存在汇市到股市的单向风险溢出而在97.5%置信水平下存在股市和汇市间的双向瞬时溢出效应。因此,监管层应制定金融稳定政策,防止两者间的风险传导。

英文摘要：

This article uses AR-GARCH model and POT model to estimate Va R of 95% and 97. 5% confidence,and employs risk spillover test based on cross-correlation function to analyse risk spillover between Hushen 300 index and RMB / US dollar exchange rate by daily data. The empirical result shows that there is risk spillover from exchange rate market to stock market under95% confidence and bidirectional transient risk spillover under 97. 5% confidence. So,regulators should make financial stability policy to avoid risk transmission between them.