Granger causality using Jacobian in neural networks

Granger causality is a commonly used method for uncovering information flow and dependencies in a time series. Here, we introduce JGC (Jacobian Granger causality), a neural network-based approach to Granger causality using the Jacobian as a measure of variable importance, and propose a variable sele...

Full description

Saved in:
Bibliographic Details
Main Authors: Suryadi, Chew, Lock Yue, Ong, Yew Soon
Other Authors: School of Physical and Mathematical Sciences
Format: Article
Language:English
Published: 2023
Subjects:
Online Access:https://hdl.handle.net/10356/165593
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-165593
record_format dspace
spelling sg-ntu-dr.10356-1655932023-04-03T15:41:45Z Granger causality using Jacobian in neural networks Suryadi Chew, Lock Yue Ong, Yew Soon School of Physical and Mathematical Sciences School of Computer Science and Engineering Engineering::Computer science and engineering Causal Modeling Target Variable Granger causality is a commonly used method for uncovering information flow and dependencies in a time series. Here, we introduce JGC (Jacobian Granger causality), a neural network-based approach to Granger causality using the Jacobian as a measure of variable importance, and propose a variable selection procedure for inferring Granger causal variables with this measure, using criteria of significance and consistency. The resulting approach performs consistently well compared to other approaches in identifying Granger causal variables, the associated time lags, as well as interaction signs. In addition, we also discuss the need for contemporaneous variables in Granger causal modeling as well as how these neural network-based approaches reduce the impact of nonseparability in dynamical systems, a problem where predictive information on a target variable is not unique to its causes, but also contained in the history of the target variable itself. Published version 2023-04-03T05:30:05Z 2023-04-03T05:30:05Z 2023 Journal Article Suryadi, Chew, L. Y. & Ong, Y. S. (2023). Granger causality using Jacobian in neural networks. Chaos: An Interdisciplinary Journal of Nonlinear Science, 33(2), 023126-. https://dx.doi.org/10.1063/5.0106666 1054-1500 https://hdl.handle.net/10356/165593 10.1063/5.0106666 36859223 2-s2.0-85148673584 2 33 023126 en Chaos: An Interdisciplinary Journal of Nonlinear Science © 2023 Author(s). All rights reserved. This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in Suryadi, Chew, L. Y. & Ong, Y. S. (2023). Granger causality using Jacobian in neural networks. Chaos: An Interdisciplinary Journal of Nonlinear Science, 33(2), 023126- and may be found at https://dx.doi.org/10.1063/5.0106666. application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Causal Modeling
Target Variable
spellingShingle Engineering::Computer science and engineering
Causal Modeling
Target Variable
Suryadi
Chew, Lock Yue
Ong, Yew Soon
Granger causality using Jacobian in neural networks
description Granger causality is a commonly used method for uncovering information flow and dependencies in a time series. Here, we introduce JGC (Jacobian Granger causality), a neural network-based approach to Granger causality using the Jacobian as a measure of variable importance, and propose a variable selection procedure for inferring Granger causal variables with this measure, using criteria of significance and consistency. The resulting approach performs consistently well compared to other approaches in identifying Granger causal variables, the associated time lags, as well as interaction signs. In addition, we also discuss the need for contemporaneous variables in Granger causal modeling as well as how these neural network-based approaches reduce the impact of nonseparability in dynamical systems, a problem where predictive information on a target variable is not unique to its causes, but also contained in the history of the target variable itself.
author2 School of Physical and Mathematical Sciences
author_facet School of Physical and Mathematical Sciences
Suryadi
Chew, Lock Yue
Ong, Yew Soon
format Article
author Suryadi
Chew, Lock Yue
Ong, Yew Soon
author_sort Suryadi
title Granger causality using Jacobian in neural networks
title_short Granger causality using Jacobian in neural networks
title_full Granger causality using Jacobian in neural networks
title_fullStr Granger causality using Jacobian in neural networks
title_full_unstemmed Granger causality using Jacobian in neural networks
title_sort granger causality using jacobian in neural networks
publishDate 2023
url https://hdl.handle.net/10356/165593
_version_ 1764208166527041536