Positivity for Gaussian graphical models

DOI10.1016/J.AAM.2013.03.001zbMATH Open1279.62135DBLPjournals/aam/DraismaST13arXiv1210.0390OpenAlexW1989745763WikidataQ57427082 ScholiaQ57427082MaRDI QIDQ394750FDOQ394750

Authors: Seth Sullivant, Kelli Talaska, Jan Draisma 

Publication date: 27 January 2014

Published in: Advances in Applied Mathematics (Search for Journal in Brave)

Abstract: Gaussian graphical models are parametric statistical models for jointly normal random variables whose dependence structure is determined by a graph. In previous work, we introduced trek separation, which gives a necessary and sufficient condition in terms of the graph for when a subdeterminant is zero for all covariance matrices that belong to the Gaussian graphical model. Here we extend this result to give explicit cancellation-free formulas for the expansions of nonzero subdeterminants.

Full work available at URL: https://arxiv.org/abs/1210.0390

Recommendations

• Trek separation for Gaussian graphical models
• Nested covariance determinants and restricted trek separation in Gaussian graphical models
• On Gaussian conditional independence structures
• Gaussian graphical models with toric vanishing ideals
• Smoothness of Gaussian conditional independence models

zbMATH Keywords

conditional independencedeterminantsLinström-Gessel-Viennot lemma

Mathematics Subject Classification ID

Multivariate analysis (62H99) Applications of graph theory (05C90) Determinants, permanents, traces, other special matrix functions (15A15)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Global identifiability of linear structural equation models
• Half-trek criterion for generic identifiability of linear structural equation models
• On a parametrization of positive semidefinite matrices with zeros
• Trek separation for Gaussian graphical models
• Binomial determinants, paths, and hook length formulae
• On the Vector Representations of Induced Matroids
• Loop-erased walks and total positivity
• A non-commutative version of Jacobi's equality on the cofactors of a matrix

Cited In (8)

• On a parametrization of positive semidefinite matrices with zeros
• Causal Structural Learning via Local Graphs
• Exact inference on Gaussian graphical models of arbitrary topology using path-sums
• Gaussian graphical models with toric vanishing ideals
• Brownian motion tree models are toric
• Partial correlation hypersurfaces in Gaussian graphical models
• Nested covariance determinants and restricted trek separation in Gaussian graphical models
• Trek separation for Gaussian graphical models

This page was built for publication: Positivity for Gaussian graphical models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q394750)