HARNet: A Convolutional Neural Network for Realized Volatility Forecasting

Center for Financial Studies Working Paper no. 680, 2022

25 Pages Posted: 31 May 2022

See all articles by Rafael Reisenhofer

Rafael Reisenhofer

University of Vienna - Faculty of Science and Mathematics

Xandro Bayer

University of Vienna - Department of Statistics and Operations Research

Nikolaus Hautsch

University of Vienna - Department of Statistics and Operations Research

Date Written: May 21, 2022

Abstract

Despite the impressive success of deep neural networks in many application areas, neural network models have so far not been widely adopted in the context of volatility forecasting. In this work, we aim to bridge the conceptual gap between established time series approaches, such as the Heterogeneous Autoregressive (HAR) model (Corsi, 2009), and state-of-the-art deep neural network models. The newly introduced HARNet is based on a hierarchy of dilated convolutional layers, which facilitates an exponential growth of the receptive field of the model in the number of model parameters. HARNets allow for an explicit initialization scheme such that before optimization, a HARNet yields identical predictions as the respective baseline HAR model. Particularly when considering the QLIKE error as a loss function, we find that this approach significantly stabilizes the optimization of HARNets. We evaluate the performance of HARNets with respect to three different stock market indexes. Based on this evaluation, we formulate clear guidelines for the optimization of HARNets and show that HARNets can substantially improve upon the forecasting accuracy of their respective HAR baseline models. In a qualitative analysis of the filter weights learnt by a HARNet, we report clear patterns regarding the predictive power of past information. Among information from the previous week, yesterday and the day before, yesterday's volatility makes by far the most contribution to today's realized volatility forecast. Moroever, within the previous month, the importance of single weeks diminishes almost linearly when moving further into the past.

Suggested Citation

Reisenhofer, Rafael and Bayer, Xandro and Hautsch, Nikolaus, HARNet: A Convolutional Neural Network for Realized Volatility Forecasting (May 21, 2022). Center for Financial Studies Working Paper no. 680, 2022, Available at SSRN: https://ssrn.com/abstract=4116642 or http://dx.doi.org/10.2139/ssrn.4116642

Rafael Reisenhofer

University of Vienna - Faculty of Science and Mathematics

Xandro Bayer

University of Vienna - Department of Statistics and Operations Research

Nikolaus Hautsch (Contact Author)

University of Vienna - Department of Statistics and Operations Research ( email )

Kolingasse 14
Vienna, A-1090
Austria

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

Paper statistics

Downloads
29
Abstract Views
122
PlumX Metrics