Towards a new viewpoint on causality for time series

Michel Fliess 1, 2, * Cédric Join 2, 3, 4
Abstract : Causation between time series is a most important topic in econometrics, financial engineering, biological and psychological sciences, and many other fields. A new setting is introduced for examining this rather abstract concept. The corresponding calculations, which are much easier than those required by the celebrated Granger-causality, do not necessitate any deterministic or probabilistic modeling. Some convincing computer simulations are presented.
Complete list of metadatas

https://hal-polytechnique.archives-ouvertes.fr/hal-00991942
Contributor : Michel Fliess <>
Submitted on : Monday, September 15, 2014 - 9:18:22 PM
Last modification on : Wednesday, March 27, 2019 - 4:41:27 PM
Long-term archiving on : Tuesday, December 16, 2014 - 11:45:42 AM

File

AEJ-causal.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00991942, version 2

Citation

Michel Fliess, Cédric Join. Towards a new viewpoint on causality for time series. Colloque Modélisation, Contrôle et Analyse des Systèmes en l'honneur du Professeur Abdelhaq EL JAI, May 2014, Ifrane, Morocco. ⟨hal-00991942v2⟩

Share

Metrics

Record views

514

Files downloads

102