Markov properties for mixed graphs

Abstract

In this paper, we unify the Markov theory of a variety of different types of graphs used in graphical Markov models by introducing the class of loopless mixed graphs, and show that all independence models induced by $m$-separation on such graphs are compositional graphoids. We focus in particular on the subclass of ribbonless graphs which as special cases include undirected graphs, bidirected graphs, and directed acyclic graphs, as well as ancestral graphs and summary graphs. We define maximality of such graphs as well as a pairwise and a global Markov property. We prove that the global and pairwise Markov properties of a maximal ribbonless graph are equivalent for any independence model that is a compositional graphoid.

Published in at http://dx.doi.org/10.3150/12-BEJ502 the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm)

References in corpus (3)

• Discrete chain graph models
• Probability distributions with summary graph structure
• Stable mixed graphs

Cited by in corpus (24)

• Nested Markov Properties for Acyclic Directed Mixed Graphs
• Markovian acyclic directed mixed graphs for discrete data
• Markov Properties for Graphical Models with Cycles and Latent Variables
• Frustration index and Cheeger inequalities for discrete and continuous magnetic Laplacians
• Causal Etiology of the Research of James M. Robins
• Alternative Markov and Causal Properties for Acyclic Directed Mixed Graphs
• Ordering-Based Causal Structure Learning in the Presence of Latent Variables
• A review of Gaussian Markov models for conditional independence
• Pairwise Markov properties for regression graphs
• Faithfulness of Probability Distributions and Graphs
• Large-Scale Local Causal Inference of Gene Regulatory Relationships
• Causal Structure Discovery from Distributions Arising from Mixtures of DAGs
• Computing Maximum Likelihood Estimates for Gaussian Graphical Models with Macaulay2
• Error AMP Chain Graphs
• Markov Properties of Discrete Determinantal Point Processes
• Graphical Markov models: overview
• Distributional Invariances and Interventional Markov Equivalence for Mixed Graph Models
• The neighborhood lattice for encoding partial correlations in a Hilbert space
• Self-adhesivity in lattices of abstract conditional independence models
• Generalized Permutohedra from Probabilistic Graphical Models
• Bell's local causality is a d-separation criterion
• Every LWF and AMP chain graph originates from a set of causal models
• Causal Effect Identification in Acyclic Directed Mixed Graphs and Gated Models
• Conditions and Assumptions for Constraint-based Causal Structure Learning