Rehabilitating Mean–Variance Portfolio Selection: Theory and Evidence
证明当资产收益服从偏斜椭圆广义位置与尺度分布时，均值-方差分析是合理的，纠正了其仅适用于正态分布的误解，并展示该分布下均值-方差与其他均值-风险方法等价，通过数值与实证研究验证其实际价值。
Recent research has proven that the application of mean–variance portfolio selection is justified if, and only if, asset returns follow a skew-elliptical generalized location and scale (SEGLS) distribution. This irrefutably corrects the widespread fallacy that mean–variance analysis can be used only for portfolios with normally or symmetrically distributed constituents. To make this important finding accessible to a wide range of academics and practitioners, the authors of this article present it in a nontechnical form and additionally highlight that, under the SEGLS distribution and some mild axiomatic requirements, mean–variance analysis and many alternative mean-risk approaches deliver the same optimal portfolios. In a numerical study, they illustrate the key features of the novel SEGLS distribution. In an empirical study, they emphasize its practical relevance by gathering existing and providing new evidence in its favor.