Shapley values for LightGBM model applied to regime detection - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year : 2021

Shapley values for LightGBM model applied to regime detection

Abstract

We consider a gradient boosting decision trees (GBDT) approach to predict large S&P 500 price drops from a set of 150 technical, fundamental and macroeconomic features. We report an improved accuracy of GBDT over other machine learning (ML) methods on the S&P 500 futures prices. We show that retaining fewer and carefully selected features provides improvements across all ML approaches. Shapley values have recently been introduced from game theory to the field of ML. They allow for a robust identification of the most important variables predicting stock market crises, and of a local explanation of the crisis probability at each date, through a consistent features attribution. We apply this methodology to analyse in detail the March 2020 financial meltdown, for which the model offered a timely out of sample prediction. This analysis unveils in particular the contrarian predictive role of the tech equity sector before and after the crash.
Fichier principal
Vignette du fichier
EXTRAAMAS_2021.pdf (749.93 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03320300 , version 1 (15-08-2021)

Identifiers

  • HAL Id : hal-03320300 , version 1

Cite

J J Ohana, S Ohana, Eric Benhamou, D Saltiel, B Guez. Shapley values for LightGBM model applied to regime detection. 2021. ⟨hal-03320300⟩
239 View
896 Download

Share

Gmail Mastodon Facebook X LinkedIn More