Reading dependencies from covariance graphs

DOI10.1016/J.IJAR.2012.06.025zbMATH Open1266.60020arXiv1010.4504OpenAlexW1888202434MaRDI QIDQ2375355FDOQ2375355

Authors: Jose M. Peña 

Publication date: 13 June 2013

Published in: International Journal of Approximate Reasoning (Search for Journal in Brave)

Abstract: The covariance graph (aka bi-directed graph) of a probability distribution $p$ is the undirected graph $G$ where two nodes are adjacent iff their corresponding random variables are marginally dependent in $p$. In this paper, we present a graphical criterion for reading dependencies from $G$, under the assumption that $p$ satisfies the graphoid properties as well as weak transitivity and composition. We prove that the graphical criterion is sound and complete in certain sense. We argue that our assumptions are not too restrictive. For instance, all the regular Gaussian probability distributions satisfy them.

Full work available at URL: https://arxiv.org/abs/1010.4504

Recommendations

• An algorithm for reading dependencies from the minimal undirected independence map of a graphoid that satisfies weak transitivity
• Gaussian covariance faithful Markov trees
• Linear dependencies represented by chain graphs. With comments and a rejoinder by the authors
• Covariance decomposition in undirected Gaussian graphical models
• Gaussian Markov distributions over finite graphs

zbMATH Keywords

chain graphsconcentration graphscovariance graphs

Mathematics Subject Classification ID

Probability distributions: general theory (60E05) Characterization and structure theory for multivariate probability distributions; copulas (62H05)

Cited In (4)

• On a Dualization of Graphical Gaussian Models: A Correction Note
• Covariance decomposition in undirected Gaussian graphical models
• Gaussian covariance faithful Markov trees
• Learning AMP chain graphs and some marginal models thereof under faithfulness

This page was built for publication: Reading dependencies from covariance graphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2375355)