基于均值-CVaR-熵的社保基金最优投资组合模型

摘要

为研究社保基金最优投资组合问题,在借鉴现代投资组合理论的基础上,从证券投资组合理论的风险度量着手,用条件在险价值(Conditional Value at Risk,CVaR)和熵来共同度量风险,提出新的风险度量模型:均值- CVaR -熵模型.在保证投资组合收益率的前提下,以CVaR和叉熵函数的线性组合为最小目标函数,在险价值(Value at Risk,VaR)为约束条件,构建考虑交易成本和政策约束下不允许卖空的基于均值- CVaR -熵的社保基金投资组合模型,探讨社保基金的投资方式及投资比例的分配问题,并利用实际数据求得该模型的最优解及各资产的分配比例.结果表明多元化投资是我国社保基金投资实现保值增值目的的必然选择.%In order to research the optimal portfolio problem of the social security fund,on the basis of modern portfolio theory,we start from the risk measurement of portfolio theory,use the conditional value at risk (CVaR) and entropy to measure risk,and propose a new risk measurement model:mean-CVaR-entropy model. In the context of guarantee portfolio return rate,using the linear combination of CVaR and cross-entropy function as the minimum objective function,value at risk (VaR) as a constraint,based on the mean-CVaR-entropy, we build the social security fund portfolio model without short sales,considering transaction costs,policy constraints,and discuss the investment of social security fund and the allocation of investment ratio. And we use actual data to obtain the optimal solution of the model and the asset allocation ratio. The results show that diversification is the inevitable choice of China's social security fund investment to achieve the purpose of conserving and increasing the value.