Group symmetry and covariance regularization

DOI10.1214/12-EJS723zbMATH Open1295.62023arXiv1111.7061MaRDI QIDQ1950873FDOQ1950873

Authors: Parikshit Shah, Venkat Chandrasekaran 

Publication date: 28 May 2013

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

Abstract: Statistical models that possess symmetry arise in diverse settings such as random fields associated to geophysical phenomena, exchangeable processes in Bayesian statistics, and cyclostationary processes in engineering. We formalize the notion of a symmetric model via group invariance. We propose projection onto a group fixed point subspace as a fundamental way of regularizing covariance matrices in the high-dimensional regime. In terms of parameters associated to the group we derive precise rates of convergence of the regularized covariance matrix and demonstrate that significant statistical gains may be expected in terms of the sample complexity. We further explore the consequences of symmetry on related model-selection problems such as the learning of sparse covariance and inverse covariance matrices. We also verify our results with simulations.

Full work available at URL: https://arxiv.org/abs/1111.7061

Recommendations

• Erratum: Group symmetry and covariance regularization
• Group Symmetric Robust Covariance Estimation
• scientific article; zbMATH DE number 7071841
• General canonical correlations with applications to group symmetry models
• Group Invariance and Convex Matrix Analysis
• Invariance of covariance structures under groups of transformations
• Gaussian field on the symmetric group: prediction and learning
• Gaussian estimates and regularized groups
• On Gaussian group convex models

zbMATH Keywords

covariance selectionexchangeabilityhigh-dimensional asymptoticsgroup invariance

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Estimation in multivariate analysis (62H12) Ordinary representations and characters (20C15)

Cites Work

• Title not available (Why is that?)
• High-dimensional graphs and variable selection with the Lasso
• High dimensional covariance matrix estimation using a factor model
• Title not available (Why is that?)
• Title not available (Why is that?)
• ON STATIONARY PROCESSES IN THE PLANE
• Covariance regularization by thresholding
• Invariant normal models
• Symmetry and lattice conditional independence in a multivariate normal distribution
• Invariant normal models with recursive graphical Markov structure.
• Regularized estimation of large covariance matrices
• Title not available (Why is that?)
• Testing and Estimation for a Circular Stationary Model
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• Local operator theory, random matrices and Banach spaces.
• First-Order Methods for Sparse Covariance Selection
• Testing Compound Symmetry in a Normal Multivariate Distribution
• Rejoinder: Latent variable graphical model selection via convex optimization
• Title not available (Why is that?)
• Title not available (Why is that?)
• Exploiting group symmetry in truss topology optimization
• Group symmetry in interior-point methods for semidefinite program
• Symmetry groups, semidefinite programs, and sums of squares
• Reduction of symmetric semidefinite programs using the regular \(\ast\)-representation
• Cyclostationarity: half a century of research
• Title not available (Why is that?)
• New Code Upper Bounds From the Terwilliger Algebra and Semidefinite Programming
• Four short stories about Toeplitz matrix calculations
• Title not available (Why is that?)
• On the estimation and testing of spatial interaction in Gaussian lattice processes
• New upper bounds for kissing numbers from semidefinite programming
• On group Fourier analysis and symmetry preserving discretizations of PDEs
• Fastest mixing Markov chain on graphs with symmetries
• Sample Criteria for Testing Equality of Means, Equality of Variances, and Equality of Covariances in a Normal Multivariate Distribution
• Title not available (Why is that?)
• A wreath product group approach to signal and image processing. I: Multiresolution analysis
• Symmetry in semidefinite programs
• Gauss-Markov random fields (CMrf) with continuous indices

Cited In (5)

• Gaussian and robust Kronecker product covariance estimation: existence and uniqueness
• Title not available (Why is that?)
• Symmetries in directed Gaussian graphical models
• Matrix compression along isogenic blocks
• Model selection in the space of Gaussian models invariant by symmetry

Uses Software

• gRc
• GMRFLib

This page was built for publication: Group symmetry and covariance regularization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1950873)