Distributionally Robust Mean-Variance Portfolio Selection with Wasserstein Distances

• 成果类型: Article
• 署名作者: Blanchet, Jose; Chen, Lin; Zhou, Xun Yu
• 署名单位: Stanford University; Columbia University
• 刊物名称: MANAGEMENT SCIENCE
• ISSN/ISSBN: 0025-1909
• DOI: 10.1287/mnsc.2021.4155
• 发表日期: 2022
• 页码: 6382-6410
• 关键词: mean-variance portfolio selection; robust model; Wasserstein Distance; robust Wasserstein profile inference

摘要

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.