A review of Gaussian Markov models for conditional independence

DOI10.1016/J.JSPI.2019.09.008zbMATH Open1437.62195OpenAlexW2975015986WikidataQ127181386 ScholiaQ127181386MaRDI QIDQ2301082FDOQ2301082

Authors: Irene Córdoba, Concha Bielza, Pedro Larrañaga 

Publication date: 28 February 2020

Published in: Journal of Statistical Planning and Inference (Search for Journal in Brave)

Abstract: Markov models lie at the interface between statistical independence in a probability distribution and graph separation properties. We review model selection and estimation in directed and undirected Markov models with Gaussian parametrization, emphasizing the main similarities and differences. These two model classes are similar but not equivalent, although they share a common intersection. We present the existing results from a historical perspective, taking into account the amount of literature existing from both the artificial intelligence and statistics research communities, where these models were originated. We cover classical topics such as maximum likelihood estimation and model selection via hypothesis testing, but also more modern approaches like regularization and Bayesian methods. We also discuss how the Markov models reviewed fit in the rich hierarchy of other, higher level Markov model classes. Finally, we close the paper overviewing relaxations of the Gaussian assumption and pointing out the main areas of application where these Markov models are nowadays used.

Full work available at URL: https://arxiv.org/abs/1606.07282

Recommendations

• scientific article; zbMATH DE number 4211299
• Gaussian Markov distributions over finite graphs
• Ancestral graph Markov models.
• Model selection for Gaussian concentration graphs
• Conditional independence and basic Markov properties

zbMATH Keywords

model selectionparameter estimationconditional independenceGaussian Markov model

Mathematics Subject Classification ID

Bayesian inference (62F15) Gaussian processes (60G15) Estimation in multivariate analysis (62H12) Measures of association (correlation, canonical correlation, etc.) (62H20) Ridge regression; shrinkage estimators (Lasso) (62J07) Research exposition (monographs, survey articles) pertaining to statistics (62-02)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• A Monte Carlo method for computing the marginal likelihood in nondecomposable Gaussian graphical models
• A SINful approach to Gaussian graphical model selection
• A Theorem about Random Fields
• A linear non-Gaussian acyclic model for causal discovery
• A note on the Lasso for Gaussian graphical model selection
• A robust procedure for Gaussian graphical model search from microarray data with \(p\) larger than \(n\)
• Alternative Markov properties for chain graphs
• An Introduction to Spatial Point Processes and Markov Random Fields
• Analogies between Multiplicative Models in Contingency Tables and Covariance Selection
• Ancestral graph Markov models.
• Asymptotic behaviour of the number of labelled essential acyclic digraphs and labelled chain graphs
• Asymptotically efficient estimation of covariance matrices with linear structure
• CAM: causal additive models, high-dimensional order search and penalized regression
• Causation, prediction, and search
• Cholesky decomposition of a hyper inverse Wishart matrix
• Comment on: Sequences of regressions and their independences
• Comments on: Sequences of regressions and their independencies
• Concave penalized estimation of sparse Gaussian Bayesian networks
• Conditional independence for statistical operations
• Counting Markov equivalence classes for DAG models on trees
• Counting and exploring sizes of Markov equivalence classes of directed acyclic graphs
• Decomposable graphical Gaussian model determination
• Decomposition of maximum likelihood in mixed graphical interaction models
• Dependence and independence
• Discrete chain graph models
• Elliptical graphical modelling
• Estimating high-dimensional directed acyclic graphs with the PC-algorithm
• Exact formulas for the normalizing constants of Wishart distributions for graphical models
• Experiments in stochastic computation for high-dimensional graphical models
• Flexible covariance estimation in graphical Gaussian models
• Fusion, propagation, and structuring in belief networks
• Gaussian graphical model estimation with false discovery rate control
• Geometry of maximum likelihood estimation in Gaussian graphical models
• Geometry of the faithfulness assumption in causal inference
• Graphical and Recursive Models for Contingency Tables
• Graphical models
• Graphical models for associations between variables, some of which are qualitative and some quantitative
• Graphical models for genetic analyses
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• High-dimensional graphs and variable selection with the Lasso
• High-dimensional learning of linear causal networks via inverse covariance estimation
• High-dimensional semiparametric Gaussian copula graphical models
• Hyper Inverse Wishart Distribution for Non-decomposable Graphs and its Application to Bayesian Inference for Gaussian Graphical Models
• Hyper Markov laws in the statistical analysis of decomposable graphical models
• Hypersurfaces and their singularities in partial correlation testing
• Independence properties of directed markov fields
• Inference in hybrid Bayesian networks using mixtures of polynomials
• Lasso-type recovery of sparse representations for high-dimensional data
• Learning local dependence in ordered data
• Learning mixtures of truncated basis functions from data
• Linear Recursive Equations, Covariance Selection, and Path Analysis
• Linear dependencies represented by chain graphs. With comments and a rejoinder by the authors
• Logical and algorithmic properties of conditional independence and graphical models
• Low-Order Conditional Independence Graphs for Inferring Genetic Networks
• Marginalization and conditioning for LWF chain graphs
• Marginalizing and conditioning in graphical models
• Markov Properties for Acyclic Directed Mixed Graphs
• Markov fields and log-linear interaction models for contingency tables
• Markov properties for mixed graphs
• Markov random fields and their applications
• Mixtures of truncated basis functions
• Model Search among Multiplicative Models
• Model selection and estimation in the Gaussian graphical model
• Model selection for Gaussian concentration graphs
• Model selection through sparse maximum likelihood estimation for multivariate Gaussian or binary data
• Multiple testing and error control in Gaussian graphical model selection
• Multivariate Correlation Models with Mixed Discrete and Continuous Variables
• Multivariate Gaussians, semidefinite matrix completion, and convex algebraic geometry
• Normal linear regression models with recursive graphical Markov structure
• Objective Bayes factors for Gaussian directed acyclic graphical models
• Objective Bayesian model selection in Gaussian graphical models
• Objective Bayesian search of Gaussian directed acyclic graphical models for ordered variables with non-local priors
• On the Non-Negative Garrotte Estimator
• On the Validity of the Markov Interpretation of Path Diagrams of Gaussian Structural Equations Systems with Correlated Errors
• On the causal interpretation of acyclic mixed graphs under multivariate normality
• On the conditions used to prove oracle results for the Lasso
• On the prior and posterior distributions used in graphical modelling
• Order-independent constraint-based causal structure learning
• PC algorithm for nonparanormal graphical models
• Parameter priors for directed acyclic graphical models and the characterization of several probability distributions
• Penalized likelihood methods for estimation of sparse high-dimensional directed acyclic graphs
• Positive definite completion problems for Bayesian networks
• Positive definite completions of partial Hermitian matrices
• Posterior graph selection and estimation consistency for high-dimensional Bayesian DAG models
• Probability distributions with summary graph structure
• Propagation of Probabilities, Means, and Variances in Mixed Graphical Association Models
• Rectangular Confidence Regions for the Means of Multivariate Normal Distributions
• Regularized rank-based estimation of high-dimensional nonparanormal graphical models
• Robust estimators for nondecomposable elliptical graphical models
• Separoids: a mathematical framework for conditional independence and irrelevance
• Sequences of regressions and their independences
• Simulation of hyper-inverse Wishart distributions for non-decomposable graphs
• Simulation of hyper-inverse Wishart distributions in graphical models
• Sparse inverse covariance estimation with the graphical lasso
• Sparse permutation invariant covariance estimation
• Sparsistency and rates of convergence in large covariance matrix estimation
• Stable mixed graphs
• Statistics for high-dimensional data. Methods, theory and applications.
• Stochastic inequalities relating a class of log-likelihood ratio statistics to their asymptotic \(\chi ^ 2\) distribution
• The Adaptive Lasso and Its Oracle Properties
• The Isserlis matrix and its application to non-decomposable graphical Gaussian models
• The nonparanormal: semiparametric estimation of high dimensional undirected graphs
• The size distribution for Markov equivalence classes of acyclic digraph models.
• Uniform consistency in causal inference
• Unifying Markov properties for graphical models
• Wishart distributions for decomposable graphs
• \(\ell_{0}\)-penalized maximum likelihood for sparse directed acyclic graphs

Cited In (4)

• A mathematical view of weights-of-evidence, conditional independence, and logistic regression in terms of Markov random fields
• Gaussian-related undirected graphical models for circular variables
• Mutual conditional independence and its applications to model selection in Markov networks
• Directional tests in Gaussian graphical models

Uses Software

• glasso
• Separoids
• TETRAD
• HdBCS

This page was built for publication: A review of Gaussian Markov models for conditional independence

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2301082)