中文摘要：

VaR和CVaR是目前两种主流风险度量工具。条件VaR和条件CVaR是基于市场风险因子在已知条件（或信息）下的分布来计量和测算VaR和CVaR,能够及时地根据变化的条件来重新估计风险进而进行有效的风险管理,是对传统的基于边际分布的VaR和CVaR指标的有益补充。另外一方面,近年来非参数核估计方法因模型设定灵活、方便处理变量相依结构等优点备受关注。在本文,我们用条件VaR和条件CVaR的非参数核估计法,对我国A股市场的风险进行测算。结果得出：条件VaR和条件CVaR能揭示出深证成指和上证综指之间的不同风险特征;条件VaR和条件CVaR的测算结果并非总是一致;系统风险估计值对已知条件的敏感性高于深发展A和万科A两只股票的个股风险。以上风险特征在边际VaR和边际CVaR下无法得到。

英文摘要：

Value-at-Risk（VaR） and Conditional Value-at-Risk（CVaR） are two main current popular risk measurement tools presently.Conditional VaR and conditional CVaR are computed with respect to conditional distribution based on known conditions（or information）.They are able to timely remeasure the risk according to the varying conditions,and conduct the effective risk management correspondingly.They are useful supplements of traditional VaR（CVaR） that are computed with respect to marginal distribution.On the other hand,nonparametric kernel estimation method got broad attention recently because its model is flexible and it can process dependent structure problem very easy.In this paper,we use nonparametric kernel estimation method of conditional VaR and conditional CVaR to measure risk in Chinese A share market.The following conclusions can be obtained：Conditional VaR and conditional CVaR can reveal the different risk features between SSE Composite Index and SZSE Component Index.The results of conditional VaR and conditional CVaR are not consistent always.The value of systematic risk is more sensitive to known conditions than individual stock risk of SDB A shares and Vanke A shares.These risk features above can＇t be measured by marginal VaR and CVaR.