Header

UZH-Logo

Maintenance Infos

Measuring Granger-causal effects in multivariate time series by system editing: PASCAL codes, executable, and toy data (human readable)


Pascual-Marqui, Roberto D; Biscay, Rolando J; Bosch-Bayard, Jorge; Faber, Pascal; Kinoshita, Toshihiko; Kochi, Kieko; Milz, Patricia; Nishida, Keiichiro; Yoshimura, Masafumi (2018). Measuring Granger-causal effects in multivariate time series by system editing: PASCAL codes, executable, and toy data (human readable). bioRxiv 1, University of Zurich.

Abstract

What is the role of each node in a system of many interconnected nodes? This can be quantified by comparing the dynamics of the nodes in the intact system, with their modified dynamics in the edited system, where one node is deleted. In detail, the spectra are calculated from a causal multivariate autoregressive model for the intact system. Next, without re-estimation, one node is deleted from the model and the modified spectra at all other nodes are re-calculated. The change in spectra from the edited system to the intact system quantifies the role of the deleted node, giving a measure of its Granger-causal effects (CFX) on the system. A generalization of this novel measure is available for networks (i.e. for groups of nodes), which quantifies the role of each network in a system of many networks. For the sake of reproducible research, program codes (PASCAL), executable file, and toy data in human readable format are included in the supplementary material.

Abstract

What is the role of each node in a system of many interconnected nodes? This can be quantified by comparing the dynamics of the nodes in the intact system, with their modified dynamics in the edited system, where one node is deleted. In detail, the spectra are calculated from a causal multivariate autoregressive model for the intact system. Next, without re-estimation, one node is deleted from the model and the modified spectra at all other nodes are re-calculated. The change in spectra from the edited system to the intact system quantifies the role of the deleted node, giving a measure of its Granger-causal effects (CFX) on the system. A generalization of this novel measure is available for networks (i.e. for groups of nodes), which quantifies the role of each network in a system of many networks. For the sake of reproducible research, program codes (PASCAL), executable file, and toy data in human readable format are included in the supplementary material.

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Downloads

18 downloads since deposited on 18 Jan 2019
15 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Working Paper
Communities & Collections:04 Faculty of Medicine > Psychiatric University Hospital Zurich > Clinic for Psychiatry, Psychotherapy, and Psychosomatics
04 Faculty of Medicine > The KEY Institute for Brain-Mind Research
Dewey Decimal Classification:610 Medicine & health
Language:English
Date:30 December 2018
Deposited On:18 Jan 2019 08:54
Last Modified:30 Jan 2020 13:46
Series Name:bioRxiv
Number of Pages:12
OA Status:Hybrid
Free access at:Publisher DOI. An embargo period may apply.
Publisher DOI:https://doi.org/10.1101/504068

Download

Hybrid Open Access

Download PDF  'Measuring Granger-causal effects in multivariate time series by system editing: PASCAL codes, executable, and toy data (human readable)'.
Preview
Content: Published Version
Language: English
Filetype: PDF
Size: 792kB
View at publisher
Licence: Creative Commons: Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)