Persistence in Factor-Based Supervised Learning Models

32 Pages Posted: 28 Oct 2021 Publication Status: Published

Multiple version iconThere are 2 versions of this paper


In this paper, we document the importance of memory in machine learning (ML)-based models relying on firm characteristics for asset pricing. We find that predictive algorithms perform best when they are trained on long samples, with long-term returns as dependent variables. In addition, we report that persistent features play a prominent role in these models. When applied to portfolio choice, we find that investors are always better off predicting annual returns, even when rebalancing at lower frequencies (monthly or quarterly). Our results remain robust to transaction costs and risk scaling, thus providing useful indications to quantitative asset managers.

Keywords: Factor Investing, Machine learning, Asset Pricing, Autocorrelation

Suggested Citation

Coqueret, Guillaume, Persistence in Factor-Based Supervised Learning Models. Available at SSRN: https://ssrn.com/abstract=3951817 or http://dx.doi.org/10.2139/ssrn.3951817

Guillaume Coqueret (Contact Author)

EMLYON Business School ( email )

23 Avenue Guy de Collongue
Ecully, 69132

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

Paper statistics

Abstract Views
PlumX Metrics