Trek separation for Gaussian graphical models

DOI10.1214/09-AOS760zbMATH Open1189.62091arXiv0812.1938OpenAlexW3099019541WikidataQ57439783 ScholiaQ57439783MaRDI QIDQ973882FDOQ973882

Seth Sullivant, Jan Draisma, Kelli Talaska

Publication date: 26 May 2010

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: Gaussian graphical models are semi-algebraic subsets of the cone of positive definite covariance matrices. Submatrices with low rank correspond to generalizations of conditional independence constraints on collections of random variables. We give a precise graph-theoretic characterization of when submatrices of the covariance matrix have small rank for a general class of mixed graphs that includes directed acyclic and undirected graphs as special cases. Our new trek separation criterion generalizes the familiar $d$-separation criterion. Proofs are based on the trek rule, the resulting matrix factorizations and classical theorems of algebraic combinatorics on the expansions of determinants of path polynomials.

Full work available at URL: https://arxiv.org/abs/0812.1938

zbMATH Keywords

Bayesian networklinear regressionconditional independencegraphical model[https://portal.mardi4nfdi.de/w/index.php?title=+Special%3ASearch&search=Gessel-Viennot-Lindstr%EF%BF%BD%EF%BF%BDm+lemma&go=Go Gessel-Viennot-Lindstr��m lemma]trek rule

Mathematics Subject Classification ID

Multivariate analysis (62H99) Linear regression; mixed models (62J05) Applications of graph theory (05C90) Characterization and structure theory for multivariate probability distributions; copulas (62H05)

Cites Work

• Title not available (Why is that?)
• Causation, prediction, and search
• Ancestral graph Markov models.
• The Method of Path Coefficients
• Binomial determinants, paths, and hook length formulae
• On the Vector Representations of Induced Matroids
• Alternative Markov properties for chain graphs
• Loop-erased walks and total positivity
• Moments of minors of Wishart matrices
• Algebraic geometry of Gaussian Bayesian networks

Cited In (23)

• Positivity for Gaussian graphical models
• Matrix Schubert varieties and Gaussian conditional independence models
• Geometry of the faithfulness assumption in causal inference
• Half-trek criterion for identifiability of latent variable models
• Identifying species network features from gene tree quartets under the coalescent model
• On the causal interpretation of acyclic mixed graphs under multivariate normality
• Causal Structural Learning via Local Graphs
• Title not available (Why is that?)
• Directed Gaussian graphical models with toric vanishing ideals
• Directed Expected Utility Networks
• Multi-Trek Separation in Linear Structural Equation Models
• Wald tests of singular hypotheses
• Gaussian graphical models with toric vanishing ideals
• Comments on: Sequences of regressions and their independencies
• Computing maximum likelihood estimates for Gaussian graphical models with Macaulay2
• Title not available (Why is that?)
• Brownian motion tree models are toric
• Half-trek criterion for generic identifiability of linear structural equation models
• Multivariate Gaussians, semidefinite matrix completion, and convex algebraic geometry
• Coherent combination of probabilistic outputs for group decision making: an algebraic approach
• Partial correlation hypersurfaces in Gaussian graphical models
• Title not available (Why is that?)
• Nested covariance determinants and restricted trek separation in Gaussian graphical models

Uses Software

• TETRAD

This page was built for publication: Trek separation for Gaussian graphical models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q973882)