Atlantis Press Paper Details: Graphical Models for the Identification of Causal Structures in Multivariate Time Series Models

back to author index

title:

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

publication:

JCIS-2006 Proceedings

part of series:

Advances in Intelligent Systems Research

ISBN:

978-90-78677-01-7

ISSN:

1951-6851

DOI:

doi:10.2991/jcis.2006.171 (how to use a DOI)

author(s):

Alessio Moneta, Peter Spirtes

corresponding author:

Alessio Moneta

publication date:

October 2006

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.

© Atlantis Press. This article is distributed under the terms of the Creative Commons Attribution License, which permits non-commercial use, distribution and reproduction in any medium, provided the original work is properly cited.

full text:

JCIS06-CIEF-28.pdf (86 K)