High-dimensional graphs and variable selection with the Lasso

DOI10.1214/009053606000000281zbMATH Open1113.62082arXivmath/0608017OpenAlexW3098834468WikidataQ105584248 ScholiaQ105584248MaRDI QIDQ2500458FDOQ2500458

Authors: Nicolai Meinshausen, Peter Bühlmann 

Publication date: 24 August 2006

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: The pattern of zero entries in the inverse covariance matrix of a multivariate normal distribution corresponds to conditional independence restrictions between variables. Covariance selection aims at estimating those structural zeros from data. We show that neighborhood selection with the Lasso is a computationally attractive alternative to standard covariance selection for sparse high-dimensional graphs. Neighborhood selection estimates the conditional independence restrictions separately for each node in the graph and is hence equivalent to variable selection for Gaussian linear models. We show that the proposed neighborhood selection scheme is consistent for sparse high-dimensional graphs. Consistency hinges on the choice of the penalty parameter. The oracle value for optimal prediction does not lead to a consistent neighborhood estimate. Controlling instead the probability of falsely joining some distinct connectivity components of the graph, consistent estimation for sparse graphs is achieved (with exponential rates), even when the number of variables grows as the number of observations raised to an arbitrary power.

Full work available at URL: https://arxiv.org/abs/math/0608017

Recommendations

• Sparse inverse covariance estimation with the graphical lasso
• The sparsity and bias of the LASSO selection in high-dimensional linear regression
• Sparse covariance thresholding for high-dimensional variable selection
• A note on the Lasso for Gaussian graphical model selection
• Efficient estimation of covariance selection models

zbMATH Keywords

covariance selectionGaussian graphical modelspenalized regression

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Multivariate analysis (62H99) Linear regression; mixed models (62J05) Applications of graph theory (05C90) Estimation in multivariate analysis (62H12) Ridge regression; shrinkage estimators (Lasso) (62J07)

Cites Work

• Introduction to Graphical Modelling
• Least angle regression. (With discussion)
• Weak convergence and empirical processes. With applications to statistics
• Title not available (Why is that?)
• Persistene in high-dimensional linear predictor-selection and the virtue of overparametrization
• Asymptotics for Lasso-type estimators.
• Title not available (Why is that?)
• Linear Model Selection by Cross-Validation
• A Statistical View of Some Chemometrics Regression Tools
• Gaussian Markov distributions over finite graphs
• Functional aggregation for nonparametric regression.
• Model selection for Gaussian concentration graphs
• Atomic decomposition by basis pursuit
• Dependency networks for inference, collaborative filtering, and data visualization
• Title not available (Why is that?)

Cited In (only showing first 100 items - show all)

• High-dimensional inference: confidence intervals, \(p\)-values and R-software \texttt{hdi}
• Change-point detection in high-dimensional covariance structure
• On stepwise pattern recovery of the fused Lasso
• Hierarchical inference for genome-wide association studies: a view on methodology with software
• High-dimensional semiparametric Gaussian copula graphical models
• Inferring sparse Gaussian graphical models with latent structure
• On the conditions used to prove oracle results for the Lasso
• Model selection through sparse maximum likelihood estimation for multivariate Gaussian or binary data
• Sparse estimation of conditional graphical models with application to gene networks
• Graphical models for zero-inflated single cell gene expression
• Group symmetry and covariance regularization
• High dimensional sparse covariance estimation via directed acyclic graphs
• Sparse permutation invariant covariance estimation
• Adaptive Lasso estimators for ultrahigh dimensional generalized linear models
• Sparse directed acyclic graphs incorporating the covariates
• Simultaneous analysis of Lasso and Dantzig selector
• Covariance-regularized regression and classification for high dimensional problems
• Robust sparse Gaussian graphical modeling
• On principal graphical models with application to gene network
• Sparse semiparametric discriminant analysis
• High-dimensional generalized linear models and the lasso
• Network exploration via the adaptive LASSO and SCAD penalties
• High-dimensional change-point estimation: combining filtering with convex optimization
• Regularized estimation of large covariance matrices
• Estimation of sparse directed acyclic graphs for multivariate counts data
• A Unified Framework for Structured Graph Learning via Spectral Constraints
• Model selection for factorial Gaussian graphical models with an application to dynamic regulatory networks
• Nearly unbiased variable selection under minimax concave penalty
• A Projection Based Conditional Dependence Measure with Applications to High-dimensional Undirected Graphical Models
• Covariate Selection in High-Dimensional Generalized Linear Models With Measurement Error
• Factor-Adjusted Regularized Model Selection
• Graph-Guided Banding of the Covariance Matrix
• Inferring multiple graphical structures
• Confidence intervals for high-dimensional inverse covariance estimation
• Stability Selection
• Asymptotic normality and optimalities in estimation of large Gaussian graphical models
• Gaussian graphical model estimation with false discovery rate control
• Honest confidence regions and optimality in high-dimensional precision matrix estimation
• Maximum Likelihood Estimation Over Directed Acyclic Gaussian Graphs
• Sure Independence Screening for Ultrahigh Dimensional Feature Space
• On asymptotically optimal confidence regions and tests for high-dimensional models
• Correlated variables in regression: clustering and sparse estimation
• Bayesian model selection approach for coloured graphical Gaussian models
• Bayesian structure learning in sparse Gaussian graphical models
• Best subset selection via a modern optimization lens
• Fast Computation of Latent Correlations
• A sparse conditional Gaussian graphical model for analysis of genetical genomics data
• Missing values: sparse inverse covariance estimation and an extension to sparse regression
• A general algorithm for covariance modeling of discrete data
• Simultaneous multiple response regression and inverse covariance matrix estimation via penalized Gaussian maximum likelihood
• Sparse inverse covariance estimation with the graphical lasso
• Multiple testing and error control in Gaussian graphical model selection
• Factor models and variable selection in high-dimensional regression analysis
• High-dimensional additive modeling
• Rates of convergence of the adaptive LASSO estimators to the oracle distribution and higher order refinements by the bootstrap
• Statistical significance in high-dimensional linear models
• Sparse recovery under matrix uncertainty
• Variable selection and regression analysis for graph-structured covariates with an application to genomics
• Confidence Intervals for Low Dimensional Parameters in High Dimensional Linear Models
• Large Covariance Estimation by Thresholding Principal Orthogonal Complements
• The sparse Laplacian shrinkage estimator for high-dimensional regression
• The log-linear group-lasso estimator and its asymptotic properties
• Boosting algorithms: regularization, prediction and model fitting
• Regularization for Cox's proportional hazards model with NP-dimensionality
• High-dimensional regression with noisy and missing data: provable guarantees with nonconvexity
• Oracle inequalities for high dimensional vector autoregressions
• Bayesian graphical models for differential pathways
• Regularized rank-based estimation of high-dimensional nonparanormal graphical models
• A unified framework for high-dimensional analysis of \(M\)-estimators with decomposable regularizers
• Sparse nonparametric graphical models
• \(\ell_{1}\)-penalization for mixture regression models
• Robust graphical modeling of gene networks using classical and alternative \(t\)-distributions
• Exact test theory in Gaussian graphical models
• Tests for Gaussian graphical models
• Lasso-type recovery of sparse representations for high-dimensional data
• Performance bounds for parameter estimates of high-dimensional linear models with correlated errors
• Quasi-likelihood and/or robust estimation in high dimensions
• Structured sparsity through convex optimization
• Lasso-driven inference in time and space
• High-dimensional variable selection
• A note on the Lasso for Gaussian graphical model selection
• The sparsity and bias of the LASSO selection in high-dimensional linear regression
• Penalised inference for lagged dependent regression in the presence of autocorrelated residuals
• High-dimensional Ising model selection using \(\ell _{1}\)-regularized logistic regression
• Least angle and \(\ell _{1}\) penalized regression: a review
• SCAD-penalized regression in high-dimensional partially linear models
• A constrained \(\ell1\) minimization approach for estimating multiple sparse Gaussian or nonparanormal graphical models
• High-dimensional simultaneous inference with the bootstrap
• Sparsistency and rates of convergence in large covariance matrix estimation
• Inferring large graphs using \(\ell_1\)-penalized likelihood
• The graphical lasso: new insights and alternatives
• ROCKET: robust confidence intervals via Kendall's tau for transelliptical graphical models
• Estimating time-varying networks
• Adaptive estimation of the baseline hazard function in the Cox model by model selection, with high-dimensional covariates
• Some sharp performance bounds for least squares regression with \(L_1\) regularization
• Hypothesis Testing of Matrix Graph Model with Application to Brain Connectivity Analysis
• Covariance estimation: the GLM and regularization perspectives
• Learning high-dimensional directed acyclic graphs with latent and selection variables
• UPS delivers optimal phase diagram in high-dimensional variable selection
• Flexible covariance estimation in graphical Gaussian models

Uses Software

• PDCO
• lars
• MIM

This page was built for publication: High-dimensional graphs and variable selection with the Lasso

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2500458)