| Issue |
ESAIM: Proc.
Volume 49, February 2015
CMACS 2014 - Colloque de Modélisation, Analyse et Contrôle des Systèmes
|
|
|---|---|---|
| Page(s) | 37 - 52 | |
| DOI | https://doi.org/10.1051/proc/201549004 | |
| Published online | 13 March 2015 | |
Towards a new viewpoint on causality for time series
1 LIX (CNRS, UMR 7161), École
polytechnique, 91128
Palaiseau,
France.
2 AL.I.E.N. (ALgèbre pour
Identification & Estimation Numériques), 24-30 rue Lionnois, BP 60120, 54003
Nancy,
France.
3 CRAN (CNRS, UMR 7039), Université de
Lorraine, BP 239,
54506
Vandœuvre-lès-Nancy,
France.
4 Projet Non-A, INRIA Lille –
Nord-Europe, France.
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.
Résumé
La causalité entre chroniques est un sujet capital en économétrie, ingénierie financière, sciences biologiques et psychologiques, et quantité d’autres domaines. On introduit ici une nouvelle approche pour traiter ce concept abstrait. Les calculs, qui sont beaucoup plus simples que ceux liés à la causalité de Granger, bien connue, ne nécessitent aucune modélisation, déterministe ou probabiliste. On présente plusieurs simulations numériques réussies.
© EDP Sciences, SMAI 2015
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.