Sparse permutation invariant covariance estimation

DOI10.1214/08-EJS176zbMATH Open1320.62135arXiv0801.4837MaRDI QIDQ1951760FDOQ1951760

Authors: Elizaveta Levina, Ji Zhu, P. J. Bickel, Adam J. Rothman 

Publication date: 24 May 2013

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: The paper proposes a method for constructing a sparse estimator for the inverse covariance (concentration) matrix in high-dimensional settings. The estimator uses a penalized normal likelihood approach and forces sparsity by using a lasso-type penalty. We establish a rate of convergence in the Frobenius norm as both data dimension $p$ and sample size $n$ are allowed to grow, and show that the rate depends explicitly on how sparse the true concentration matrix is. We also show that a correlation-based version of the method exhibits better rates in the operator norm. We also derive a fast iterative algorithm for computing the estimator, which relies on the popular Cholesky decomposition of the inverse but produces a permutation-invariant estimator. The method is compared to other estimators on simulated data and on a real data example of tumor tissue classification using gene expression data.

Full work available at URL: https://arxiv.org/abs/0801.4837

Recommendations

• High-dimensional covariance estimation based on Gaussian graphical models
• A well-conditioned and sparse estimation of covariance and inverse covariance matrices using a joint penalty
• A constrained \(\ell _{1}\) minimization approach to sparse precision matrix estimation
• Sparse estimation of a covariance matrix
• Sparsistency and rates of convergence in large covariance matrix estimation

zbMATH Keywords

Lassosparsitycovariance matrixCholesky decompositionlarge \(p\) small \(n\)high dimension low sample size

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Measures of association (correlation, canonical correlation, etc.) (62H20)

Cites Work

• Pathwise coordinate optimization
• Title not available (Why is that?)
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• High-dimensional graphs and variable selection with the Lasso
• High dimensional covariance matrix estimation using a factor model
• Covariance regularization by thresholding
• Sparsistency and rates of convergence in large covariance matrix estimation
• A well-conditioned estimator for large-dimensional covariance matrices
• Sparse inverse covariance estimation with the graphical lasso
• Some theory for Fisher's linear discriminant function, `naive Bayes', and some alternatives when there are many more variables than observations
• On the distribution of the largest eigenvalue in principal components analysis
• Regularized estimation of large covariance matrices
• Variable selection using MM algorithms
• Estimating high-dimensional directed acyclic graphs with the PC-algorithm
• Title not available (Why is that?)
• Model selection and estimation in the Gaussian graphical model
• Parsimonious Covariance Matrix Estimation for Longitudinal Data
• Asymptotics of sample eigenstructure for a large dimensional spiked covariance model
• Sparse estimation of large covariance matrices via a nested Lasso penalty
• First-Order Methods for Sparse Covariance Selection
• Operator norm consistent estimation of large-dimensional sparse covariance matrices
• Nonparametric estimation of large covariance matrices of longitudinal data
• Title not available (Why is that?)
• Title not available (Why is that?)
• Estimation of a covariance matrix under Stein's loss
• Empirical Bayes estimation of the multivariate normal covariance matrix
• Covariance matrix selection and estimation via penalised normal likelihood
• Nonlinear Programming
• A SINful approach to Gaussian graphical model selection
• Estimation of a covariance matrix with zeros
• Efficient estimation of covariance selection models
• Estimation of high-dimensional prior and posterior covariance matrices in Kalman filter vari\-ants

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

• Prediction in abundant high-dimensional linear regression
• Estimating time-varying networks
• High-dimensional semiparametric Gaussian copula graphical models
• Sparse estimation of conditional graphical models with application to gene networks
• Covariance estimation: the GLM and regularization perspectives
• Transposable regularized covariance models with an application to missing data imputation
• Optimal rates of convergence for covariance matrix estimation
• High dimensional sparse covariance estimation via directed acyclic graphs
• Scaling it up: stochastic search structure learning in graphical models
• A constrained \(\ell _{1}\) minimization approach to sparse precision matrix estimation
• Likelihood-based selection and sharp parameter estimation
• Layer-wise learning strategy for nonparametric tensor product smoothing spline regression and graphical models
• Sparse directed acyclic graphs incorporating the covariates
• Covariance-regularized regression and classification for high dimensional problems
• Robust sparse Gaussian graphical modeling
• Discussion: Latent variable graphical model selection via convex optimization
• Bootstrap inference for network construction with an application to a breast cancer microarray study
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• Network exploration via the adaptive LASSO and SCAD penalties
• Structural pursuit over multiple undirected graphs
• Efficient estimation of approximate factor models via penalized maximum likelihood
• Simultaneous inference for pairwise graphical models with generalized score matching
• Rejoinder: Latent variable graphical model selection via convex optimization
• Time varying undirected graphs
• Improved multivariate normal mean estimation with unknown covariance when \(p\) is greater than \(n\)
• Discussion: Latent variable graphical model selection via convex optimization
• A multiple testing approach to the regularisation of large sample correlation matrices
• A two-step method for estimating high-dimensional Gaussian graphical models
• Gaussian and robust Kronecker product covariance estimation: existence and uniqueness
• Gemini: graph estimation with matrix variate normal instances
• Estimation of sparse directed acyclic graphs for multivariate counts data
• Variable selection of linear programming discriminant estimator
• Structure estimation for discrete graphical models: generalized covariance matrices and their inverses
• Linear estimating equations for exponential families with application to Gaussian linear concentration models
• Covariance and precision matrix estimation for high-dimensional time series
• Penalized estimation in high-dimensional hidden Markov models with state-specific graphical models
• A computationally fast alternative to cross-validation in penalized Gaussian graphical models
• Joint estimation of precision matrices in heterogeneous populations
• High dimensional inverse covariance matrix estimation via linear programming
• Posterior convergence rates for estimating large precision matrices using graphical models
• Random matrix theory in statistics: a review
• Fast and adaptive sparse precision matrix estimation in high dimensions
• Regularization for stationary multivariate time series
• Adaptive covariance matrix estimation through block thresholding
• Adaptive estimation of stationary Gaussian fields
• An Expectation Conditional Maximization Approach for Gaussian Graphical Models
• A well-conditioned and sparse estimation of covariance and inverse covariance matrices using a joint penalty
• Model selection for factorial Gaussian graphical models with an application to dynamic regulatory networks
• A joint convex penalty for inverse covariance matrix estimation
• 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
• 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
• A sparse Ising model with covariates
• High-dimensional regression with noisy and missing data: provable guarantees with nonconvexity
• Latent variable graphical model selection via convex optimization
• Design-free estimation of variance matrices
• Sparse Matrix Graphical Models
• Regularized rank-based estimation of high-dimensional nonparanormal graphical models
• Estimating sufficient reductions of the predictors in abundant high-dimensional regressions
• A unified framework for high-dimensional analysis of \(M\)-estimators with decomposable regularizers
• Sparse nonparametric graphical models
• High-dimensionality effects in the Markowitz problem and other quadratic programs with linear constraints: risk underestimation
• Graphical group ridge
• High-dimensional covariance matrix estimation with missing observations
• Estimation of (near) low-rank matrices with noise and high-dimensional scaling
• Estimating sparse precision matrix: optimal rates of convergence and adaptive estimation
• Covariance regularization by thresholding
• High-dimensional Ising model selection using \(\ell _{1}\)-regularized logistic regression
• Optimal computational and statistical rates of convergence for sparse nonconvex learning problems
• A constrained \(\ell1\) minimization approach for estimating multiple sparse Gaussian or nonparanormal graphical models
• Sparsistency and rates of convergence in large covariance matrix estimation
• Variable selection in multivariate linear models with high-dimensional covariance matrix estimation
• The graphical lasso: new insights and alternatives
• Transfer Learning in Large-Scale Gaussian Graphical Models with False Discovery Rate Control
• Graph-guided banding of the covariance matrix
• Sparsistency and agnostic inference in sparse PCA
• Estimating spatial covariance using penalised likelihood with weightedL₁penalty
• ROCKET: robust confidence intervals via Kendall's tau for transelliptical graphical models
• Spectral clustering via sparse graph structure learning with application to proteomic signaling networks in cancer
• Scalable Bayesian matrix normal graphical models for brain functional networks
• A self-calibrated direct approach to precision matrix estimation and linear discriminant analysis in high dimensions
• Ensemble sparse estimation of covariance structure for exploring genetic disease data
• Promote sign consistency in the joint estimation of precision matrices
• Block-diagonal precision matrix regularization for ultra-high dimensional data
• Sparse overlapped linear discriminant analysis
• Sparse estimation of high-dimensional correlation matrices
• Gaussian and bootstrap approximations for high-dimensional U-statistics and their applications
• Estimating Time-Varying Graphical Models
• The cluster graphical Lasso for improved estimation of Gaussian graphical models
• Adjusted regularization in latent graphical models: application to multiple-neuron spike count data
• Adjusted regularization of cortical covariance
• Discussion: Latent variable graphical model selection via convex optimization
• Monitoring the covariance matrix with fewer observations than variables
• Consistency of large dimensional sample covariance matrix under weak dependence
• Generalized symmetric ADMM for separable convex optimization

Uses Software

• pcalg
• glasso

This page was built for publication: Sparse permutation invariant covariance estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1951760)