投资过程中的非Markowitz风险偏好分析及最优选择

Non-Markowitz Risk-Preference Analysis and Optimal Choice During the Investment Process

Markowitz最优投资组合理论,是建立在投资者传统理性假设(或称为Markowitz风险偏好假设)之上的。但是在现实世界中,这一假设很难实现。为此,本文首先基于CVaR模型建立了非Markowitz风险偏好的度量方法;然后,利用无差异曲线和信用评级思想将投资者的风险偏好水平划分成了五种风险偏好等级,进而给出了各风险偏好等级下最优投资组合的确定方法。作为应用,本文对我国金融市场的最优投资组合进行了实证考察,结果发现:风险偏好程度越高,投资组合的绩效越好,投资者对损失率也更为敏感;此外,相比于传统的Markowitz风险偏好假设,非Markowitz风险偏好假设下的投资组合绩效更好。

The Markowitz optimal portfolio theory is based on the investor’s traditional rational hypothesis(Markowitz risk-preference hypothesis). But in the real world, Markowitz risk-preference hypothesis is hard to achieve. Therefore, based on the CVaR model, we first build a Non-Markowitz risk-preference measurement method. Moreover, the risk-preference level of investors is divided into five grades by using the idea of indifference curve and credit rating, and then the method of determining the optimal portfolio under different risk preference grades is given. Using the methods above, we empirically analyze the optimal portfolio of China’s financial market. The result shows that the higher the risk-preference level is, the better the portfolio performs and the more sensitive the investor is to loss rate. In addition, compared to the traditional Markowitz risk-preference hypothesis, the portfolio performs much better under the Non-Markowitz risk-preference hypothesis.