摘要
研究了基于CVaR约束的的最优投资决策问题,为避免维数障碍,针对Fredrik提出的CVaR投资组合优化线性规划模型还原为非线性规划。通过引入缩进因子,改进PSO算法,使粒子在迭代过程中保持在可行域内。最后,通过算例证明了该文方法的有效性,计算结果表明,投资组合优化后的损失期望收益率、标准差、受险价值、条件受险价值等重要风险衡量指标都有了较大改进。
In this paper,we research the portfolio optimization based on CVaR.In order to avoid the drawback of dimension obstacle,a portfolio optimization linear program model based on CVaR,which is proposed by Fredrik,is reverted to a nonlinear one.Guaranteed by the indentation factor,the improved PSO algorithm can keep the iteration particles in feasible region.Finally,we prove the effectiveness of this method by a numerical example.The results show that the optimized portfolios＇ expected yield rate,standard deviation,VaR and CVaR obviously decrease.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2010年第1期179-182,191,共5页
Journal of Wuhan University of Technology
关键词
CVAR
投资组合优化
PSO算法
CVaR
pmffolio optimization
particle swarm optimization algorithm
