Multiple Testing and Error Control in Gaussian Graphical Model Selection

Abstract

Graphical models provide a framework for exploration of multivariate dependence patterns. The connection between graph and statistical model is made by identifying the vertices of the graph with the observed variables and translating the pattern of edges in the graph into a pattern of conditional independences that is imposed on the variables' joint distribution. Focusing on Gaussian models, we review classical graphical models. For these models the defining conditional independences are equivalent to vanishing of certain (partial) correlation coefficients associated with individual edges that are absent from the graph. Hence, Gaussian graphical model selection can be performed by multiple testing of hypotheses about vanishing (partial) correlation coefficients. We show and exemplify how this approach allows one to perform model selection while controlling error rates for incorrect edge inclusion.

Published in at http://dx.doi.org/10.1214/088342307000000113 the Statistical Science (http://www.imstat.org/sts/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (7)

• High-dimensional graphs and variable selection with the Lasso
• Generalizations of the Familywise Error Rate
• Control of generalized error rates in multiple testing
• Estimation of a Covariance Matrix with Zeros
• A Mutagenetic Tree Hidden Markov Model for Longitudinal Clonal HIV Sequence Data
• Maximum Likelihood Estimation in Gaussian Chain Graph Models under the Alternative Markov Property
• Perfect Tree-Like Markovian Distributions

Cited by in corpus (26)

• Regularized rank-based estimation of high-dimensional nonparanormal graphical models
• Robust graphical modeling of gene networks using classical and alternative T-distributions
• Inferring sparse Gaussian graphical models with latent structure
• Local and Global Inference for High Dimensional Nonparanormal Graphical Models
• On Semiparametric Exponential Family Graphical Models
• Estimation of Gaussian graphs by model selection
• Modeling dependent gene expression
• Direct Estimation of Differences in Causal Graphs
• False Discovery Rate Control for High-Dimensional Networks of Quantile Associations Conditioning on Covariates
• A review of Gaussian Markov models for conditional independence
• Learning directed acyclic graphs based on sparsest permutations
• Graphical Models for Discrete and Continuous Data
• Graph inference with clustering and false discovery rate control
• GGM knockoff filter: False Discovery Rate Control for Gaussian Graphical Models
• The Nonparanormal: Semiparametric Estimation of High Dimensional Undirected Graphs
• Bayesian Graph Selection Consistency Under Model Misspecification
• Joint Estimation of Multiple Graphical Models from High Dimensional Time Series
• Family-wise error rate control in Gaussian graphical model selection via Distributionally Robust Optimization
• Optimal statistical decision for Gaussian graphical model selection
• Asymptotic control of FWER under Gaussian assumption: application to correlation tests
• False Discovery Rates in Biological Networks
• High-dimensional inference on covariance structures via the extended cross-data-matrix methodology
• Graphical Exponential Screening
• Comparison of statistical procedures for Gaussian graphical model selection
• On a parametrization of positive semidefinite matrices with zeros
• Some Statistical Problems with High Dimensional Financial data