Machine Learning for Predicting Stock Return Volatility

63 Pages Posted: 30 Dec 2021

See all articles by Damir Filipović

Damir Filipović

Ecole Polytechnique Fédérale de Lausanne; Swiss Finance Institute

Amir Khalilzadeh

Ecole Polytechnique Fédérale de Lausanne

Date Written: December 23, 2021

Abstract

We use machine learning methods to predict stock return volatility. Our out-of-sample prediction of realised volatility for a large cross-section of US stocks over the sample period from 1992 to 2016 is on average 44.1% against the actual realised volatility of 43.8% with an R2 being as high as double the ones reported in the literature. We further show that machine learning methods can capture the stylized facts about volatility without relying on any assumption about the distribution of stock returns. Finally, we show that our long short-term memory model outperforms other models by properly carrying information from the past predictor values.

Keywords: Volatility Prediction, Volatility Clustering, LSTM, Neural Networks, Regression Trees.

JEL Classification: C51, C52, C53, C58, G17.

Suggested Citation

Filipovic, Damir and Khalilzadeh, Amir, Machine Learning for Predicting Stock Return Volatility (December 23, 2021). Swiss Finance Institute Research Paper No. 21-95, Available at SSRN: https://ssrn.com/abstract=3995529 or http://dx.doi.org/10.2139/ssrn.3995529

Damir Filipovic (Contact Author)

Ecole Polytechnique Fédérale de Lausanne ( email )

Odyssea
Station 5
Lausanne, 1015
Switzerland

HOME PAGE: http://people.epfl.ch/damir.filipovic

Swiss Finance Institute

c/o University of Geneva
40, Bd du Pont-d'Arve
CH-1211 Geneva 4
Switzerland

Amir Khalilzadeh

Ecole Polytechnique Fédérale de Lausanne ( email )

EPFL Innovation Park
Extension School
CH-1015 Lausanne, Vaud
Switzerland

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

Paper statistics

Downloads
1,612
Abstract Views
4,236
rank
16,645
PlumX Metrics