基于CVaR的多目标投资组合模型  

The Multi-objective Portfolio Model Based on CVaR

作　　者：杨天山[1] 韦增欣[1] 雷震[1] 万腾飞[1] 

机构地区：[1]广西大学数学与信息科学学院,广西南宁530004

出　　处：《数学的实践与认识》2015年第2期82-88,共7页Mathematics in Practice and Theory

基　　金：国家自然科学基金(11161003)

摘　　要：金融市场的发展与完善,以及人民收入水平的提高,使越来越多人关注金融投资并成为热点.理性的投资者总是期望风险尽可能低同时收益又尽可能高,而且希望投资的资产易于管理和管理成本低.考虑投资者多个目标的要求,将运用CVaR风险度量方法,提出一个均值—CVaR—资产数目的多目标投资组合模型,并利用多目标粒子群算法对模型进行实证分析,验证新模型的可行性和有效性,为热衷投资的投资者进行投资组合提供一个新方法.With the development of financial market and the increase of ordinary people＇s income, more and more people are concerning about the financial investment which finally becomes a hot topic. Reasonable investors are always expecting lower risks while at the same time obtaining higher earnings as well as manageable assets and lower costs on such management. In order to meet the investors＇ multiple demands, this paper will take the CVaR risk measurement by promoting a tri-objective investment model, namely, the mean value, CVaR, and the amount of assets. The Multi-objective Particle Swarm Optimization （MPSO） will be used to empirically analyze the model in order to verify its feasibility and validity. All these will offer a new way for those investors who prefer investment portfolio.

关 键 词：投资组合 多目标模型 多目标粒子群算法 CVAR 

分 类 号：F832.48[经济管理—金融学] O221.6[理学—运筹学与控制论] F224[理学—数学]