On the causal interpretation of acyclic mixed graphs under multivariate normality

DOI10.1016/J.LAA.2014.02.032zbMATH Open1314.05085arXiv1308.3728OpenAlexW2963810577WikidataQ57566386 ScholiaQ57566386MaRDI QIDQ2341881FDOQ2341881

Authors: Andreas Käufl, Mathias Drton, Christopher Fox 

Publication date: 6 May 2015

Published in: Linear Algebra and its Applications (Search for Journal in Brave)

Abstract: In multivariate statistics, acyclic mixed graphs with directed and bidirected edges are widely used for compact representation of dependence structures that can arise in the presence of hidden (i.e., latent or unobserved) variables. Indeed, under multivariate normality, every mixed graph corresponds to a set of covariance matrices that contains as a full-dimensional subset the covariance matrices associated with a causally interpretable acyclic digraph. This digraph generally has some of its nodes corresponding to hidden variables. We seek to clarify for which mixed graphs there exists an acyclic digraph whose hidden variable model coincides with the mixed graph model. Restricting to the tractable setting of chain graphs and multivariate normality, we show that decomposability of the bidirected part of the chain graph is necessary and sufficient for equality between the mixed graph model and some hidden variable model given by an acyclic digraph.

Full work available at URL: https://arxiv.org/abs/1308.3728

Recommendations

• Markovian acyclic directed mixed graphs for discrete data
• scientific article; zbMATH DE number 802837
• The hidden life of latent variables: Bayesian learning with mixed graph models
• Nested covariance determinants and restricted trek separation in Gaussian graphical models
• Markov Properties for Acyclic Directed Mixed Graphs

zbMATH Keywords

multivariate normal distributionstructural equation modelcovariance matrixgraphical modellatent variablehidden variable

Mathematics Subject Classification ID

Characterization and structure theory for multivariate probability distributions; copulas (62H05) Directed graphs (digraphs), tournaments (05C20)

Cites Work

• Causation, prediction, and search
• Title not available (Why is that?)
• Title not available (Why is that?)
• Causality. Models, reasoning, and inference
• Global identifiability of linear structural equation models
• Ancestral graph Markov models.
• Computing maximum likelihood estimates in recursive linear models with correlated errors
• On a parametrization of positive semidefinite matrices with zeros
• The Method of Path Coefficients
• Probability distributions with summary graph structure
• Lectures on algebraic statistics
• Trek separation for Gaussian graphical models
• Learning high-dimensional directed acyclic graphs with latent and selection variables
• Marginalizing and conditioning in graphical models
• The hidden life of latent variables: Bayesian learning with mixed graph models
• On the Validity of the Markov Interpretation of Path Diagrams of Gaussian Structural Equations Systems with Correlated Errors
• Markov equivalence for ancestral graphs
• A user's guide to measure theoretic probability
• Sequences of regressions and their independences
• Discrete chain graph models
• Graphical methods for efficient likelihood inference in Gaussian covariance models
• Title not available (Why is that?)
• Joint Response Graphs and Separation Induced by Triangular Systems
• On the Markov equivalence of maximal ancestral graphs

Cited In (6)

• Margins of discrete Bayesian networks
• Signed Directed Acyclic Graphs for Causal Inference
• Graphs for margins of Bayesian networks
• The hidden life of latent variables: Bayesian learning with mixed graph models
• A review of Gaussian Markov models for conditional independence
• Structural factor equation models for causal network construction via directed acyclic mixed graphs

Uses Software

• TETRAD

This page was built for publication: On the causal interpretation of acyclic mixed graphs under multivariate normality

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2341881)