A Machine Learning Approach to the Fama-French Three- and Five-Factor Models

28 Pages Posted: 24 Aug 2019

See all articles by Boubacar Diallo

Boubacar Diallo

Qatar University - College of Business and Economics

Aliyu Bagudu

AiFi Technologies LLC

Qi Zhang

People’s Bank of China

Date Written: August 21, 2019

Abstract

This research proposes new estimations of the Fama-French three- and five-factor models via a machine learning approach. Speci fically, it uses a Bayesian optimization-support vector regression (BSVR) approach to obtain predictions of portfolio returns. On data from fi ve industries' portfolio returns in the United States over the period July 1926 to January 2019, the BSVR models perform well. Our new model, called the Fama-French BSVR three-factor model, outperformed the Fama-French BSVR five-factor model. More precisely, the Fama-French BSVR three-factor estimations attain out-of-sample (testing dataset) correlation coefficients of 94% for portfolio returns for the consumption and manufacturing industries. A correlation of 92% between the predicted and experimental values of portfolio returns was found for the high-tech industry; 91% was found for the mining, construction, transportation, hotels, entertainment and finance industries. However, for the Fama-French BSVR five-factor model, the correlation coefficients lie between 48% (health industry) and 89% (high-tech industry).

Keywords: asset pricing model, factor model, machine learning, support vector regression, Bayesian optimization

JEL Classification: G12, C8, C88, C63

Suggested Citation

Diallo, Boubacar and Bagudu, Aliyu and Zhang, Qi, A Machine Learning Approach to the Fama-French Three- and Five-Factor Models (August 21, 2019). Available at SSRN: https://ssrn.com/abstract=3440840 or http://dx.doi.org/10.2139/ssrn.3440840

Boubacar Diallo (Contact Author)

Qatar University - College of Business and Economics ( email )

2713 Doha
Qatar

Aliyu Bagudu

AiFi Technologies LLC ( email )

United Arab Emirates

Qi Zhang

People’s Bank of China ( email )

China

Register to save articles to
your library

Register

Paper statistics

Downloads
13
Abstract Views
327
PlumX Metrics