Maximizing Portfolio Predictability with Machine Learning

40 Pages Posted: 17 Nov 2023

See all articles by Michael Pinelis

Michael Pinelis

Department of Economics, Cornell University

D. Ruppert

Cornell University

Date Written: November 3, 2023

Abstract

We construct the maximally predictable portfolio (MPP) of stocks using machine learning. Solving for the optimal constrained weights in the multi-asset MPP gives portfolios with a high monthly coefficient of determination, given the sample covariance matrix of predicted return errors from a machine learning model. Various models for the covariance matrix are tested. The MPPs of S&P 500 index constituents with estimated returns from Elastic Net, Random Forest, and Support Vector Regression models can outperform or underperform the index depending on the time period. Portfolios that take advantage of the high predictability of the MPP's returns and employ a Kelly criterion style strategy consistently outperform the benchmark.

Keywords: Maximally Predictable Portfolio, Machine Learning, Convex Portfolio Optimization, Empirical Stock Pricing

JEL Classification: G11, G12, C13

Suggested Citation

Pinelis, Michael and Ruppert, D., Maximizing Portfolio Predictability with Machine Learning (November 3, 2023). Available at SSRN: https://ssrn.com/abstract=4622042 or http://dx.doi.org/10.2139/ssrn.4622042

Michael Pinelis (Contact Author)

Department of Economics, Cornell University ( email )

Ithaca, NY
United States

D. Ruppert

Cornell University ( email )

School of Operations Research and Industrial Engineering
Ithaca, NY 14853
United States
607-255-9136 (Phone)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
465
Abstract Views
1,095
Rank
118,348
PlumX Metrics