Forecasting Realized Volatility: An Automatic System Using Many Features and Many Machine Learning Algorithms

62 Pages Posted: 31 Jan 2021 Last revised: 24 May 2021

See all articles by Sophia Zhengzi Li

Sophia Zhengzi Li

Rutgers, The State University of New Jersey - Rutgers Business School at Newark & New Brunswick

Yushan Tang

Rutgers, The State University of New Jersey - Rutgers Business School at Newark & New Brunswick

Date Written: May 23, 2021

Abstract

We propose an automatic machine-learning system to forecast realized volatility for S&P 100 stocks using 118 features and five machine learning algorithms. A simple average ensemble model combining all learning algorithms delivers extraordinary performance across forecast horizons, and the improvement in out-of-sample R2's translates into nontrivial economic gains. We further augment the feature set by including firm characteristics and pure noise terms, and find that the system continues to perform well after including weak or noisy features. Finally, we demonstrate that our learning system is scalable to a broader S&P 500 stock universe via hyperparameter transfer learning for nonlinear models.

Keywords: Automation, Volatility Forecasting, Machine Learning, High-Frequency Data, Realized Variance, Transfer Learning

JEL Classification: C13, C14, C52, C53, C55, C58

Suggested Citation

Li, Sophia Zhengzi and Tang, Yushan, Forecasting Realized Volatility: An Automatic System Using Many Features and Many Machine Learning Algorithms (May 23, 2021). Available at SSRN: https://ssrn.com/abstract=3776915 or http://dx.doi.org/10.2139/ssrn.3776915

Sophia Zhengzi Li (Contact Author)

Rutgers, The State University of New Jersey - Rutgers Business School at Newark & New Brunswick ( email )

100 Rockafeller Rd
Piscataway, NJ 08854
United States

Yushan Tang

Rutgers, The State University of New Jersey - Rutgers Business School at Newark & New Brunswick ( email )

1 WASHINGTON PARK
ROOM 1107W
Newark, NJ 07102
United States

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

Paper statistics

Downloads
329
Abstract Views
1,250
rank
118,590
PlumX Metrics