Graphical Models for the Identification of Causal Structures in Multivariate Time Series Models

DOI

10.2991/jcis.2006.171How to use a DOI?

Keywords

graphical models, causality, problem of identification, vector autoregressions, dynamic factor models

Abstract

In this paper we present a semi-automated search procedure to deal with the problem of the identification of the contemporaneous causal structure connected to a large class of multivariate time series models. We propose to use graphical causal models for recovering partial information about the contemporaneous causal structure of the data generating process starting from statistical properties (partial correlations) of the data. Our method permit the exclusion of a large set of causal structures which are not consistent with some statistical properties, under the assumption that any causal structure among random variables is tied to a particular configuration of partial correlations over the same random variables.

Copyright

© 2006, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Alessio Moneta
AU  - Peter Spirtes
PY  - 2006/10
DA  - 2006/10
TI  - Graphical Models for the Identification of Causal Structures in Multivariate Time Series Models
BT  - Proceedings of the 9th Joint International Conference on Information Sciences (JCIS-06)
PB  - Atlantis Press
SP  - 613
EP  - 616
SN  - 1951-6851
UR  - https://doi.org/10.2991/jcis.2006.171
DO  - 10.2991/jcis.2006.171
ID  - Moneta2006/10
ER  -