Interpretable Supervised Portfolios

30 Pages Posted: 30 Sep 2022

Date Written: September 27, 2022

Abstract

The supervised portfolios approach is an effective asset allocation strategy that engineers optimal weights before feeding them to a supervised learning algorithm. Yet, supervised learning algorithms are often seen as opaque, which undermines trust in those models, thereby limiting their adoption. To alleviate this issue, we apply an enhanced version of RuleFit, an intrinsic interpretable algorithm, which transforms a black-box non-linear predictive algorithm into a simple combination of rules. It fits a sparse linear model that includes feature interactions, derived from decision tree ensembles. Our first empirical analysis illustrates that the interpretable approach is statistically as accurate as gradient boosting on three different investment universes. Our second analysis highlights which characteristics and interactions matter for an equity portfolio manager.

Keywords: Interpretable Machine Learning, Rulefit, Supervised learning,portfolio construction

JEL Classification: G12, C62, D81

Suggested Citation

CHEVALIER, Guillaume and Coqueret, Guillaume and Raffinot, Thomas, Interpretable Supervised Portfolios (September 27, 2022). Available at SSRN: https://ssrn.com/abstract=4230955 or http://dx.doi.org/10.2139/ssrn.4230955

Guillaume CHEVALIER

AXA Investment Managers ( email )

Tour Majunga
6 Place de la Pyramide
La Défense, Paris 92908
France

Guillaume Coqueret

EMLYON Business School ( email )

23 Avenue Guy de Collongue
Ecully, 69132
France

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
798
Abstract Views
2,214
Rank
58,090
PlumX Metrics