条件风险价值（CVaR）在投资组合理论中的应用研究

【作者】 巩前锦；

【导师】 刘亚铮；

【作者基本信息】 中南大学 ， 技术经济及管理， 2003， 硕士

【摘要】 CVaR(Conditional Value-at-Risk，条件风险价值、)风险测量方法是在VaR(Value-at-Risk，风险价值)风险测量方法的缺陷基础之上所产生的，最早是在1999年底由Rockafellar提出的，其含义是：组合损失超过VaR的条件均值，反映超额损失的平均水平。它较之于VaR风险测量方法，更能体现投资组合的潜在风险。 CVaR风险测量方法有多方面的应用，如信用风险的测量、内部风险资本金的确定、资本配置、金融监管等，本文重点研究的是CVaR在投资组合理论中的运用，在研究的过程中，力求运用系统理论、归纳演绎、比较与实证分析等研究方法。首先，从总体上介绍了风险测量方法的发展过程，分析了它们的缺陷，然后再对CVaR风险测量方法进行深入的研究，对其概念、参数选择、计算、性质等方面都作了较详细的探讨，得出的结论是CVaR风险测量方法比传统的风险测量方法拥有更多的优点。其次，对CVaR在投资组合理论中的运用进行了深入的研究，这也是本文的创新之所在，解决了两个问题，一个是均值—CVaR的边界与有效前沿的问题；另一个是基于CVaR约束的最优投资组合选择问题，并对这两个问题都进行了实证研究，实质上，这两个问题的意义都是利用CVaR来指导投资决策，最终的目的是一样，不同的只是方法。再次，简单地阐述了CVaR在我国应用所面临的问题，并提出一些建议。最后，得出结论是，CVaR风险测量方法对投资决策是有指导意义的，基于CVaR约束的最优投资组合选择模型是均值—CVaR模型的有益补充。更多还原

【Abstract】 The risk measure method of Conditional Value-at-Risk is developed on basis of the shortcoming of Value-at-Risk method, which is raised by Rockafellar at 1999. The implication of CVaR is the conditional loss over VaR of portfolio, which reflects the average exceed quota. CVaR method reflects underlying loss better than VaR method.CVaR has many applications, such as measuring credit risk, deciding inner risk capital, asset allocation, banking supervision et al. This paper focuses the applications of CVaR in the portfolio theory. In the course of researching, I do my best to use the methods of systems theory, induction and deduction, comparison and empirical analysis et al. Firstly, the paper totally introduces all kinds of traditional risk measure methods and then analyses the defects of these methods, then puts forward CVaR risk measure method, introduces and analyses the definition, the parameter selecting, the calculation, the properties and applications and so on. The conclusion is that CVaR is better than the traditional methods and CVaR method is a coherent risk measure method. Secondly, the paper studies the application of CVaR in portfolio theory, which is also the new ideas of this paper. One is the bound and efficient frontier of mean-CVaR, the other is the optimal selection of portfolio on based of CVaR constraint. And using empirical analysis proved these two problems. In fact the implications of these two problems are identical. They all use CVaR to guide invest decisions. Thirdly, the paper simply analyses the problems of the use of CVaR in China and puts forwards some suggestions. Finally, concluding this paper and the conclusion is that CVaR is a better risk measure method and it can guide investment decisions, and the model of CVaR constraint is a good supplement of the model of mean-CVaR.更多还原