A review of Gaussian Markov models for conditional independence

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.

arXiv commentFix author signature

References in corpus (34)

• High-dimensional graphs and variable selection with the Lasso
• On the conditions used to prove oracle results for the Lasso
• Estimating high-dimensional directed acyclic graphs with the PC-algorithm
• Sparse permutation invariant covariance estimation
• Lasso-type recovery of sparse representations for high-dimensional data
• Sparsistency and rates of convergence in large covariance matrix estimation
• CAM: Causal additive models, high-dimensional order search and penalized regression
• Strong Completeness and Faithfulness in Bayesian Networks
• Regularized rank-based estimation of high-dimensional nonparanormal graphical models
• Geometry of the faithfulness assumption in causal inference
• Causal Discovery with Continuous Additive Noise Models
• Directed Cyclic Graphical Representations of Feedback Models
• Multiple Testing and Error Control in Gaussian Graphical Model Selection
• ${\ell }_0$-penalized maximum likelihood for sparse directed acyclic graphs
• Wishart distributions for decomposable graphs
• Parameter Priors for Directed Acyclic Graphical Models and the Characterization of Several Probability Distributions
• High-dimensional learning of linear causal networks via inverse covariance estimation
• Geometry of maximum likelihood estimation in Gaussian graphical models
• Discrete chain graph models
• Flexible covariance estimation in graphical Gaussian models
• Markov properties for mixed graphs
• Nested Markov Properties for Acyclic Directed Mixed Graphs
• Elliptical graphical modelling
• Concave Penalized Estimation of Sparse Gaussian Bayesian Networks
• Probability distributions with summary graph structure
• Learning Directed Acyclic Graphs with Penalized Neighbourhood Regression
• Stable mixed graphs
• Learning Local Dependence In Ordered Data
• High dimensional Bayesian inference for Gaussian directed acyclic graph models
• Hypersurfaces and their singularities in partial correlation testing
• Alternative Markov and Causal Properties for Acyclic Directed Mixed Graphs
• Marginalization and Conditioning for LWF Chain Graphs
• Robust estimators for non-decomposable elliptical graphical models
• Graphical Markov models, unifying results and their interpretation

Cited by in corpus (1)

• Sparse Cholesky covariance parametrization for recovering latent structure in ordered data