基于RR-EP模型的风险量化技术

经典的均值-方差投资组合模型作为现代投资组合理论的基础, 采用方差作为风险度量, 忽略了投资组合收益的非对称性. 半方差模型只侧重于风险部分的度量, 而对于高出期望值的部分 (超额收益)未予考虑. 文章针对方差、半方差方法作为风险测度的不足, 提出了一种新的度量投资风险和收益的RR-EP模型, 该模型不仅能度量投资的相对风险而且还能够对于高出期望值的那部分收益予于充分考虑. 并结合连续型随机变量的情形, 进一步讨论RR-EP模型的性质及其与传统度量风险方法的一致性. 实例分析说明了方法的实用性和有效性.

Risk measure method for investment based on RR-EP model

The classical Markowitz's mean-variance model in modern investment science uses variance as risk measure while it ignores the asymmetry of the return distribution. Semivariance method only pays attention to the measurement of risk part. It ignores the profit over expectation (the excess profit). This paper identifies the shortcoming of variance and semivariance methods and introduces a new risk measure model for investment RR-EP model. This model can not only measure the relative risk of investment but also give enough consideration to the profits over expectation. As to continuous random variables, the paper gives some properties of RR-EP model and the consistency between RR-EP model and traditional risk measure model. Case analysis proves the practicality and efficiency of this new method.