The geometry of Gaussian double Markovian distributions

Abstract

Gaussian double Markovian models consist of covariance matrices constrained by a pair of graphs specifying zeros simultaneously in the covariance matrix and its inverse. We study the semi-algebraic geometry of these models, in particular their dimension, smoothness and connectedness as well as algebraic and combinatorial properties.

arXiv comment33 pages, 1 figure. v4: added missing cases in Proposition 3.31

References in corpus (10)

• Likelihood ratio tests and singularities
• Conditional Information Inequalities for Entropic and Almost Entropic Points
• Lovasz-Saks-Schrijver ideals and coordinate sections of determinantal varieties
• Multigraded Commutative Algebra of Graph Decompositions
• The Geometry of Gaussoids
• Positive margins and primary decomposition
• Smoothness of marginal log-linear parameterizations
• Gaussoids are two-antecedental approximations of Gaussian conditional independence structures
• Construction Methods for Gaussoids
• A New Algorithm for Maximum Likelihood Estimation in Gaussian Graphical Models for Marginal Independence