基于Monte Carlo模拟和混合整数规划的CVaR(VaR)投资组合优化

Portfolio Optimization with Conditional Value-at-Risk(VaR) Based on Monte Carlo and Mixed-integer Programming

条件风险值(CVaR)也称为平均超额损失或者尾部VaR,是一致性的风险度量.基于Rock-afeller和Uryasev的CVaR投资组合理论,结合MonteCarlo模拟法和分枝定界法,建立了CVaR最优投资组合模型.以上证50指数样本股为研究对象,对中国股票市场投资组合进行了实证分析,并与经典的均方差模型及VaR模型作了比较分析.结果表明,研究模型及方法是有效的.

Conditional VaR is also called mean excess loss or tail VaR, and CVaR is a coherent measurement of risk. Based on portfolio optimization theory of CVaR put forward by Rockafeller and Uryasev, and combined with Monte Carlo simulation and branch and bound algorithm, a portfolio optimization model of CVaR is established. The stocks which composed shangzheng 50 index are selected to compose an optimal portfolio under CVaR, mean-variance and VaR measurement. Experiential analysis and comparative experiment have finally shown that the model established in this paper and the method were efficient.