稳健均值-方差模型的构建及比较研究

Construction and Comparative Research of Robust Mean-Variance Model

传统均值-方差模型容易受到离群值的影响,导致计算结果与实际情况存在偏差。针对这一情况,文章将稳健统计的思想与传统均值-方差模型相结合,构建出稳健均值-方差模型,以达到减少或消除离群值对模型计算结果影响的目的,并进行了模拟和实证分析。结果表明:当数据中不存在离群值时,使用传统和稳健均值-方差模型进行投资决策的效果基本保持一致;当数据中存在离群值时,传统均值-方差模型容易受到离群值的影响,而稳健均值-方差模型对离群值有较好的抵御能力,可获得较优的投资组合前沿和投资权重。

The traditional mean-variance model is easily affected by outliers, which leads to the deviation between the calculated results and the actual situation. In view of this reality, the paper combines the idea of robust statistics with the traditional mean-variance model to construct a robust mean-variance model, so as to achieve the aim of reducing or eliminating the influence of outliers on the calculation results of the model. The paper also makes simulation and empirical analysis. The results show that when there is no outlier in the data, the results of investment decision using traditional and robust mean-variance models are basically consistent, and that when there are outliers in the data, the traditional mean-variance model is easily affected by outliers,but the robust mean-variance model has better resistance to outliers and can obtain better portfolio frontier and investment weight.