Distributionally Robust Mean-Variance Portfolio Selection with Wasserstein Distances
Distributionally Robust Mean-Variance Portfolio Selection with Wasserstein Distances
复制标题
DOI：
10.1287/mnsc.2021.4155
复制
发表时间：
2022-09-01
影响因子：
5.4
通讯作者：
Zhou, Xun Yu
中科院分区：
文献类型：
--
作者：
Blanchet, Jose;Chen, Lin;Zhou, Xun Yu
我们重新审视Markowitz的均值-方差投资组合选择模型，考虑一个分布稳健的版本，其中分布不确定性的区域是围绕经验措施和概率措施之间的差异是由theWasserstein距离。我们把这个问题归结为一个带有额外正则化项的经验方差最小化问题。此外，我们将最近开发的推理方法扩展到我们的环境中，以便以数据驱动的方式选择分布不确定性的大小以及相关的稳健目标回报率。最后，我们报告了广泛的回测结果，标准普尔500指数，比较我们的模型与包括法玛-法国和布莱克-利特曼模型在内的几个知名模型的性能。
We revisit Markowitz's mean-variance portfolio selection model by considering a distributionally robust version, in which the region of distributional uncertainty is around the empirical measure and the discrepancy between probability measures is dictated by theWasserstein distance. We reduce this problem into an empirical variance minimization problem with an additional regularization term. Moreover, we extend the recently developed inference methodology to our setting in order to select the size of the distributional uncertainty as well as the associated robust target return rate in a data-driven way. Finally, we report extensive back-testing results on S&P 500 that compare the performance of our model with those of severalwell-knownmodels including the Fama-French and Black-Litterman models.