Stochastic Nonlinear Time Series Forecasting Using Time-Delay Reservoir Computers: Performance and Universality

Neural Networks, 55C, 59-71 (2014)

24 Pages Posted: 6 Nov 2013 Last revised: 13 May 2014

See all articles by Lyudmila Grigoryeva

Lyudmila Grigoryeva

University of Konstanz

Julie Henriques

Tea-Cegos Deployment

Laurent Larger

University of Burgundy Franche-Comté

Juan-Pablo Ortega

Centre National de la Recherche Scientifique (CNRS); Nanyang Technological University

Date Written: November 5, 2013

Abstract

Reservoir computing is a recently introduced machine learning paradigm that has already shown excellent performances in the processing of empirical data. We study a particular kind of reservoir computers called time-delay reservoirs that are constructed out of the sampling of the solution of a time-delay differential equation and show their good performance in the forecasting of the conditional covariances associated to multivariate discrete-time nonlinear stochastic processes of VEC-GARCH type as well as in the prediction of factual daily market realized volatilities computed with intraday quotes, using as training input daily log-return series of moderate size. We tackle some problems associated to the lack of task-universality for individually operating reservoirs and propose a solution based on the use of parallel arrays of time-delay reservoirs.

Keywords: Reservoir computing, echo state networks, neural computing, time-delay reservoir, time series forecasting, universality, VEC-GARCH model, volatility forecasting, realized volatility, parallel reservoir computing

JEL Classification: C45, C53, C63

Suggested Citation

Grigoryeva, Lyudmila and Henriques, Julie and Larger, Laurent and Ortega, Juan-Pablo and Ortega, Juan-Pablo, Stochastic Nonlinear Time Series Forecasting Using Time-Delay Reservoir Computers: Performance and Universality (November 5, 2013). Neural Networks, 55C, 59-71 (2014), Available at SSRN: https://ssrn.com/abstract=2350331 or http://dx.doi.org/10.2139/ssrn.2350331

Lyudmila Grigoryeva

University of Konstanz ( email )

Fach D-144
Universitätsstraße 10
Konstanz, D-78457
Germany

Julie Henriques

Tea-Cegos Deployment ( email )

11 rue Denis Papin
Besancon, F-25000
France

Laurent Larger

University of Burgundy Franche-Comté ( email )

1 rue Claude Goudimel
25030 Besancon cedex, DOUBS 25000
France

Juan-Pablo Ortega (Contact Author)

Centre National de la Recherche Scientifique (CNRS) ( email )

16 route de Gray
Besançon, 25030
France

HOME PAGE: http://juan-pablo-ortega.com

Nanyang Technological University ( email )

21 Nanyang Link
Singapore, 637371
Singapore

HOME PAGE: http://https://juan-pablo-ortega.com

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

Paper statistics

Downloads
299
Abstract Views
1,380
rank
130,618
PlumX Metrics