摘要
在随机变量为非完全分布信息的情况下,以最坏情况的条件风险价值WCVaR代替CVaR为风险度量指标,对整个投资过程进行风险控制,并考虑到市场的不完备性,如不允许卖空以及交易费用的情况,建立均值-WCVaR模型。依据沪深证券市场的实际数据,运用遗传算法进行数值仿真,实验表明该模型能更有效地控制风险。
引文
[1]Markowitz H.Portfolio Selection[J].The Journal of Finance,1952,7.
[2]Philippe J.Value at Risk[M].New York:The Mc Graw-Hill Compa nies,Inc,1997.
[3]Duffie D,Pan J.An Overview of Value at Risk[J].Journal of Deriva tives,1997,24(4).
[4]Xin-Li Zhang,Ke-Cun Zhang.Using Genetic Algorithm to Solve aNew Multi-period Stochastic Optimization Model[J].Journal of Compu tational and Applied Mathematics,2009,231.
[5]张昕丽,张可村,贾鹏.一种基于GARCH模型的多阶段随机最优组合模型[J].山西大学学报(自然科学版),2008,31(3).
[6]Zhou S S,Fukushima M.Worse-case Conditional Value-at-risk withApplication to Robust Portfolio Management[C].Working Paper,De partment of Applied Mathematics and Physics,Kyoto University,2006.
[7]Rockafeller T,Uryasev S.Optimization of Conditional VaR[J].Journalof Risk,2000,25(3).
[8]Lobo M S,Boyd S.The Worse-Case Risk of a Portfolio[J].Journal ofEconomic Dynamics and Control,2000,26(3).
[9]徐宗本.计算智能—模拟进化计算[M].北京:高等教育出版社,2004.
[10]刘小茂,李楚霖,王建华.风险资产组合的均值-CVaR有效前沿(I)[J].管理工程学报,2003,(1).
[11]金博轶.基于Copula的投资组合均值-CVaR有效前沿分析[J].统计与决策,2010,(2).