Department of Economics and Business Economics

Andreas Roepstorff

Inferring causality from noisy time series data: A test of Convergent Cross-Mapping

Research output: Contribution to book/anthology/report/proceedingArticle in proceedingsResearchpeer-review

Standard

Inferring causality from noisy time series data : A test of Convergent Cross-Mapping. / Mønster, Dan; Fusaroli, Riccardo; Tylén, Kristian; Roepstorff, Andreas; Sherson, Jacob.

Proceedings of the 1st International Conference on Complex Information Systems. ed. / Víctor Méndez Muñoz; Oleg Gusikhin; Victor Chang. SCITEPRESS Digital Library, 2016. p. 48-56.

Research output: Contribution to book/anthology/report/proceedingArticle in proceedingsResearchpeer-review

Harvard

Mønster, D, Fusaroli, R, Tylén, K, Roepstorff, A & Sherson, J 2016, Inferring causality from noisy time series data: A test of Convergent Cross-Mapping. in V Méndez Muñoz, O Gusikhin & V Chang (eds), Proceedings of the 1st International Conference on Complex Information Systems. SCITEPRESS Digital Library, pp. 48-56, 1st International Conference on Complex Information Systems, Rome, Italy, 22/04/2016. https://doi.org/10.5220/0005932600480056

APA

Mønster, D., Fusaroli, R., Tylén, K., Roepstorff, A., & Sherson, J. (2016). Inferring causality from noisy time series data: A test of Convergent Cross-Mapping. In V. Méndez Muñoz, O. Gusikhin, & V. Chang (Eds.), Proceedings of the 1st International Conference on Complex Information Systems (pp. 48-56). SCITEPRESS Digital Library. https://doi.org/10.5220/0005932600480056

CBE

Mønster D, Fusaroli R, Tylén K, Roepstorff A, Sherson J. 2016. Inferring causality from noisy time series data: A test of Convergent Cross-Mapping. Méndez Muñoz V, Gusikhin O, Chang V, editors. In Proceedings of the 1st International Conference on Complex Information Systems. SCITEPRESS Digital Library. pp. 48-56. https://doi.org/10.5220/0005932600480056

MLA

Mønster, Dan et al. "Inferring causality from noisy time series data: A test of Convergent Cross-Mapping"., Méndez Muñoz, Víctor Gusikhin, Oleg Chang, Victor (editors). Proceedings of the 1st International Conference on Complex Information Systems. SCITEPRESS Digital Library. 2016, 48-56. https://doi.org/10.5220/0005932600480056

Vancouver

Mønster D, Fusaroli R, Tylén K, Roepstorff A, Sherson J. Inferring causality from noisy time series data: A test of Convergent Cross-Mapping. In Méndez Muñoz V, Gusikhin O, Chang V, editors, Proceedings of the 1st International Conference on Complex Information Systems. SCITEPRESS Digital Library. 2016. p. 48-56 https://doi.org/10.5220/0005932600480056

Author

Mønster, Dan ; Fusaroli, Riccardo ; Tylén, Kristian ; Roepstorff, Andreas ; Sherson, Jacob. / Inferring causality from noisy time series data : A test of Convergent Cross-Mapping. Proceedings of the 1st International Conference on Complex Information Systems. editor / Víctor Méndez Muñoz ; Oleg Gusikhin ; Victor Chang. SCITEPRESS Digital Library, 2016. pp. 48-56

Bibtex

@inproceedings{68e4e6d8fa904cbe83e18385d84d4284,
title = "Inferring causality from noisy time series data: A test of Convergent Cross-Mapping",
abstract = "Convergent Cross-Mapping (CCM) has shown high potential to perform causal inference in the absence of models. We assess the strengths and weaknesses of the method by varying coupling strength and noise levels in coupled logistic maps. We find that CCM fails to infer accurate coupling strength and even causality direction in synchronized time-series and in the presence of intermediate coupling. We find that the presence of noise deterministically reduces the level of cross-mapping fidelity, while the convergence rate exhibits higher levels of robustness. Finally, we propose that controlled noise injections in intermediate-to-strongly coupled systems could enable more accurate causal inferences. Given the inherent noisy nature of real-world systems, our findings enable a more accurate evaluation of CCM applicability and advance suggestions on how to overcome its weaknesses.",
author = "Dan M{\o}nster and Riccardo Fusaroli and Kristian Tyl{\'e}n and Andreas Roepstorff and Jacob Sherson",
year = "2016",
doi = "10.5220/0005932600480056",
language = "English",
isbn = "978-989-758-181-6",
pages = "48--56",
editor = "{M{\'e}ndez Mu{\~n}oz}, V{\'i}ctor and Oleg Gusikhin and Victor Chang",
booktitle = "Proceedings of the 1st International Conference on Complex Information Systems",
publisher = "SCITEPRESS Digital Library",
note = "1st International Conference on Complex Information Systems, COMPLEXIS 2016 ; Conference date: 22-04-2016 Through 24-04-2016",
url = "http://www.complexis.org/?y=2016",

}

RIS

TY - GEN

T1 - Inferring causality from noisy time series data

T2 - 1st International Conference on Complex Information Systems

AU - Mønster, Dan

AU - Fusaroli, Riccardo

AU - Tylén, Kristian

AU - Roepstorff, Andreas

AU - Sherson, Jacob

N1 - Conference code: 1

PY - 2016

Y1 - 2016

N2 - Convergent Cross-Mapping (CCM) has shown high potential to perform causal inference in the absence of models. We assess the strengths and weaknesses of the method by varying coupling strength and noise levels in coupled logistic maps. We find that CCM fails to infer accurate coupling strength and even causality direction in synchronized time-series and in the presence of intermediate coupling. We find that the presence of noise deterministically reduces the level of cross-mapping fidelity, while the convergence rate exhibits higher levels of robustness. Finally, we propose that controlled noise injections in intermediate-to-strongly coupled systems could enable more accurate causal inferences. Given the inherent noisy nature of real-world systems, our findings enable a more accurate evaluation of CCM applicability and advance suggestions on how to overcome its weaknesses.

AB - Convergent Cross-Mapping (CCM) has shown high potential to perform causal inference in the absence of models. We assess the strengths and weaknesses of the method by varying coupling strength and noise levels in coupled logistic maps. We find that CCM fails to infer accurate coupling strength and even causality direction in synchronized time-series and in the presence of intermediate coupling. We find that the presence of noise deterministically reduces the level of cross-mapping fidelity, while the convergence rate exhibits higher levels of robustness. Finally, we propose that controlled noise injections in intermediate-to-strongly coupled systems could enable more accurate causal inferences. Given the inherent noisy nature of real-world systems, our findings enable a more accurate evaluation of CCM applicability and advance suggestions on how to overcome its weaknesses.

U2 - 10.5220/0005932600480056

DO - 10.5220/0005932600480056

M3 - Article in proceedings

SN - 978-989-758-181-6

SP - 48

EP - 56

BT - Proceedings of the 1st International Conference on Complex Information Systems

A2 - Méndez Muñoz, Víctor

A2 - Gusikhin, Oleg

A2 - Chang, Victor

PB - SCITEPRESS Digital Library

Y2 - 22 April 2016 through 24 April 2016

ER -