风险价值(VaR)和条件风险价值(CVaR)是目前两大主流风险度量工具,如何准确地对它们进行估计是风险管理实践中首要而核心的问题。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,我们在核估计的框架内讨论VaR和CVaR估计量的性质;给出投资组合VaR和CVaR对组合头寸的一阶导数向量和二阶导数矩阵的核估计公式,并用它们来讨论组合VaR和CVaR对组合头寸的敏感性和凸性。最后,我们利用中国外汇市场的实际数据做实证分析。
Value-at-Risk (VaR) and Conditional Value-at-Risk (CVaR) are two mainly popular risk measurement tools presently. How these risk measure indicators can be estimated accurately is the primary and central problem in the risk management practice. Nonparametric kernel estimation method has received broad attention recently because it can process dependent structure problem very easy and its mode] is flex- ible. In this paper, the kernel estimator of VaR and CVaR is investigated and firstly some properties of kernel estimator of VaR and CVaR are discussed. Then the investment portfolio's VaR and CVaR are defi- nited and the analytical expression is derived for the first and second derivatives of the VaR and CVaR, which are used to analyze the sensitivity and convexity of VaR and CVaR. Finally, the kernel estimation method is used to estimate the sensitivity and convexity of VaR and CVaR.