On generating random Gaussian graphical models
From MaRDI portal
Abstract: Structure learning methods for covariance and concentration graphs are often validated on synthetic models, usually obtained by randomly generating: (i) an undirected graph, and (ii) a compatible symmetric positive definite (SPD) matrix. In order to ensure positive definiteness in (ii), a dominant diagonal is usually imposed. In this work we investigate different methods to generate random symmetric positive definite matrices with undirected graphical constraints. We show that if the graph is chordal it is possible to sample uniformly from the set of correlation matrices compatible with the graph, while for general undirected graphs we rely on a partial orthogonalization method.
Recommendations
- Gaussian graphical models
- On robust Gaussian graphical modeling
- Model selection and estimation in the Gaussian graphical model
- scientific article; zbMATH DE number 922429
- Estimation of Gaussian graphs by model selection
- An Expectation Conditional Maximization Approach for Gaussian Graphical Models
- Maximum likelihood for Gaussians on graphs
- Probabilistic graphical models
- Graphical models
Cites work
- scientific article; zbMATH DE number 1134987 (Why is no real title available?)
- scientific article; zbMATH DE number 922429 (Why is no real title available?)
- A constrained \(\ell _{1}\) minimization approach to sparse precision matrix estimation
- Cholesky decomposition of a hyper inverse Wishart matrix
- Condition-Number-Regularized Covariance Estimation
- Conditional independence for statistical operations
- Covariance chains
- Generating random correlation matrices based on partial correlations
- Generating random correlation matrices based on vines and extended onion method
- Hyper Markov laws in the statistical analysis of decomposable graphical models
- Joint mean-covariance models with applications to longitudinal data: unconstrained parameterisation
- Linear Recursive Equations, Covariance Selection, and Path Analysis
- Linear dependencies represented by chain graphs. With comments and a rejoinder by the authors
- Model selection and estimation in the Gaussian graphical model
- Nonlinear shrinkage estimation of large-dimensional covariance matrices
- Regularized rank-based estimation of high-dimensional nonparanormal graphical models
- Sampling from a Manifold
- Schur products and matrix completions
- Simplicial faces of the set of correlation matrices
- Simulation of hyper-inverse Wishart distributions in graphical models
- Wishart distributions for decomposable covariance graph models
- Wishart distributions for decomposable graphs
Cited in
(8)- scientific article; zbMATH DE number 2014740 (Why is no real title available?)
- scientific article; zbMATH DE number 5957391 (Why is no real title available?)
- A Metropolis-Hastings based method for sampling from the \(G\)-Wishart distribution in Gaussian graphical models
- A Monte Carlo method for computing the marginal likelihood in nondecomposable Gaussian graphical models
- Scalable and exact sampling method for probabilistic generative graph models
- Joint density of correlations in the correlation matrix with chordal sparsity patterns
- Generation of positiv definite matrices with constraints to validate network algorithms for micro array data.
- gmat
This page was built for publication: On generating random Gaussian graphical models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q135242)
