A Machine Learning Efficient Frontier

Clark, Brian J.; Feinstein, Zachary; Simaan, Majeed

doi:10.2139/ssrn.3541387

A Machine Learning Efficient Frontier

Operations Research Letters, DOI: https://doi.org/10.1016/j.orl.2020.07.016

14 Pages Posted: 14 Mar 2020 Last revised: 7 Aug 2020

See all articles by Brian J. Clark

Zachary Feinstein

Stevens Institute of Technology - School of Business

Majeed Simaan

Stevens Institute of Technology - School of Business

Date Written: May 21, 2020

Abstract

We propose a simple approach to bridge between portfolio theory and machine learning. The outcome is an out-of-sample machine learning efficient frontier based on two assets, high risk and low risk. By rotating between the two assets, we show that the proposed frontier dominates the mean-variance efficient frontier out-of-sample. Our results, therefore, shed important light on the appeal of machine learning into portfolio selection under estimation risk.

Keywords: Portfolio Theory, Machine Learning, Tactical Asset Allocation, Estimation Risk

JEL Classification: C4, C6, C13, D8, G1

Suggested Citation: Suggested Citation

Clark, Brian J. and Feinstein, Zachary and Simaan, Majeed, A Machine Learning Efficient Frontier (May 21, 2020). Operations Research Letters, DOI: https://doi.org/10.1016/j.orl.2020.07.016, Available at SSRN: https://ssrn.com/abstract=3541387 or http://dx.doi.org/10.2139/ssrn.3541387