Causal structure learning from time series: Large regression coefficients may predict causal links better in practice than small p-values
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Causal structure learning from time series: Large regression coefficients may predict causal links better in practice than small p-values. / Weichwald, Sebastian; Emil Jakobsen, Martin; Mogensen, Phillip Bredahl; Petersen, Lasse; Thams, Nikolaj Theodor; Varando, Gherardo.
Proceedings of the NeurIPS 2019 Competition and Demonstration Track. PMLR, 2020. p. 27-36 (Proceedings of Machine Learning Research, Vol. 123).Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Causal structure learning from time series: Large regression coefficients may predict causal links better in practice than small p-values
AU - Weichwald, Sebastian
AU - Emil Jakobsen, Martin
AU - Mogensen, Phillip Bredahl
AU - Petersen, Lasse
AU - Thams, Nikolaj Theodor
AU - Varando, Gherardo
PY - 2020
Y1 - 2020
N2 - In this article, we describe the algorithms for causal structure learning from time series data that won the Causality 4 Climate competition at the Conference on Neural Information Processing Systems 2019 (NeurIPS). We examine how our combination of established ideas achieves competitive performance on semi-realistic and realistic time series data exhibiting common challenges in real-world Earth sciences data. In particular, we discuss a) a rationale for leveraging linear methods to identify causal links in non-linear systems, b) a simulation-backed explanation as to why large regression coefficients may predict causal links better in practice than small p-values and thus why normalising the data may sometimes hinder causal structure learning. For benchmark usage, we detail the algorithms here and provide implementations at {https://github.com/sweichwald/tidybench}. We propose the presented competition-proven methods for baseline benchmark comparisons to guide the development of novel algorithms for structure learning from time series.
AB - In this article, we describe the algorithms for causal structure learning from time series data that won the Causality 4 Climate competition at the Conference on Neural Information Processing Systems 2019 (NeurIPS). We examine how our combination of established ideas achieves competitive performance on semi-realistic and realistic time series data exhibiting common challenges in real-world Earth sciences data. In particular, we discuss a) a rationale for leveraging linear methods to identify causal links in non-linear systems, b) a simulation-backed explanation as to why large regression coefficients may predict causal links better in practice than small p-values and thus why normalising the data may sometimes hinder causal structure learning. For benchmark usage, we detail the algorithms here and provide implementations at {https://github.com/sweichwald/tidybench}. We propose the presented competition-proven methods for baseline benchmark comparisons to guide the development of novel algorithms for structure learning from time series.
M3 - Article in proceedings
T3 - Proceedings of Machine Learning Research
SP - 27
EP - 36
BT - Proceedings of the NeurIPS 2019 Competition and Demonstration Track
PB - PMLR
T2 - Neural Information Processing Systems Conference 2019,
Y2 - 8 December 2019 through 14 December 2019
ER -
ID: 248233218