Gaussian and robust Kronecker product covariance estimation: existence and uniqueness

DOI10.1016/J.JMVA.2016.04.001zbMATH Open1341.62132arXiv1512.00336OpenAlexW2287893571MaRDI QIDQ290708FDOQ290708

Authors: I. Soloveychik, Dmitry Trushin 

Publication date: 3 June 2016

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Abstract: We study the Gaussian and robust covariance estimation, assuming the true covariance matrix to be a Kronecker product of two lower dimensional square matrices. In both settings we define the estimators as solutions to the constrained maximum likelihood programs. In the robust case, we consider Tyler's estimator defined as the maximum likelihood estimator of a certain distribution on a sphere. We develop tight sufficient conditions for the existence and uniqueness of the estimates and show that in the Gaussian scenario with the unknown mean, $p/q+q/p+2$ samples are almost surely enough to guarantee the existence and uniqueness, where $p$ and $q$ are the dimensions of the Kronecker product factors. In the robust case with the known mean, the corresponding sufficient number of samples is $max[p/q,q/p]+1$.

Full work available at URL: https://arxiv.org/abs/1512.00336

Recommendations

• Existence and uniqueness of the maximum likelihood estimator for models with a Kronecker product covariance structure
• Existence and uniqueness of the Kronecker covariance MLE
• A note on necessary and sufficient conditions of existence and uniqueness for the maximum likelihood estimator of a Kronecker-product variance-covariance matrix
• Covariance estimation via sparse Kronecker structures
• Models with a Kronecker product covariance structure: estimation and testing

zbMATH Keywords

robust estimationconstrained covariance estimationhigh-dimensional estimationKronecker product structure

Mathematics Subject Classification ID

Point estimation (62F10) Asymptotic properties of parametric estimators (62F12) Multivariate analysis (62H99) Estimation in multivariate analysis (62H12) Parametric inference under constraints (62F30) Robustness and adaptive procedures (parametric inference) (62F35)

Cites Work

• The mle algorithm for the matrix normal distribution
• High dimensional covariance matrix estimation using a factor model
• Robust Estimation of a Location Parameter
• A distribution-free M-estimator of multivariate scatter
• Robust m-estimators of multivariate location and scatter
• Sparse inverse covariance estimation with the graphical lasso
• Group symmetry and covariance regularization
• Sparse permutation invariant covariance estimation
• Regularized estimation of large covariance matrices
• Model selection through sparse maximum likelihood estimation for multivariate Gaussian or binary data
• Model selection and estimation in the Gaussian graphical model
• Title not available (Why is that?)
• Group Symmetric Robust Covariance Estimation
• Statistics for Spatial Data
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• Some matrix-variate distribution theory: Notational considerations and a Bayesian application
• High-dimensional covariance matrix estimation with missing observations
• The likelihood ratio test for a separable covariance matrix
• On Estimation of Covariance Matrices With Kronecker Product Structure
• Geodesic Convexity and Covariance Estimation
• Existence and uniqueness of the maximum likelihood estimator for models with a Kronecker product covariance structure
• Transposable regularized covariance models with an application to missing data imputation
• Quantitative estimates of the convergence of the empirical covariance matrix in log-concave ensembles
• Models with a Kronecker product covariance structure: estimation and testing
• Geodesic convexity in nonlinear optimization
• Complex Elliptically Symmetric Distributions: Survey, New Results and Applications
• Hub Discovery in Partial Correlation Graphs
• Modeling the spatial and temporal dependence in fMRI data
• Model selection and estimation in the matrix normal graphical model
• Matrices with banded inverses: inversion algorithms and factorization of Gauss-Markov processes
• Time-Varying Autoregressive (TVAR) Models for Multiple Radar Observations
• Covariance Structure Maximum-Likelihood Estimates in Compound Gaussian Noise: Existence and Algorithm Analysis
• Knowledge-Aided Covariance Matrix Estimation and Adaptive Detection in Compound-Gaussian Noise
• Robust Shrinkage Estimation of High-Dimensional Covariance Matrices
• Unified Framework to Regularized Covariance Estimation in Scaled Gaussian Models
• On Convergence of Kronecker Graphical Lasso Algorithms
• Time Varying Autoregressive Moving Average Models for Covariance Estimation
• Cramér-Rao Lower Bounds on Covariance Matrix Estimation for Complex Elliptically Symmetric Distributions
• Tyler's Covariance Matrix Estimator in Elliptical Models With Convex Structure
• Performance Analysis of Tyler's Covariance Estimator
• Regularized Robust Estimation of Mean and Covariance Matrix Under Heavy-Tailed Distributions
• Rejoinder
• Block matrices with L-block-banded inverse: inversion algorithms
• Structured Robust Covariance Estimation

Cited In (11)

• Discrepancy between structured matrices in the power analysis of a separability test
• Existence and uniqueness of the maximum likelihood estimator for models with a Kronecker product covariance structure
• Classification With the Matrix-Variate-t Distribution
• Title not available (Why is that?)
• A review of Tyler's shape matrix and its extensions
• Maximum likelihood estimation for tensor normal models via castling transforms
• Almost sure uniqueness of a global minimum without convexity
• Rational maximum likelihood estimators of Kronecker covariance matrices
• Robust tests for scatter separability beyond Gaussianity
• Existence and uniqueness of the Kronecker covariance MLE
• A note on necessary and sufficient conditions of existence and uniqueness for the maximum likelihood estimator of a Kronecker-product variance-covariance matrix

Uses Software

• glasso

This page was built for publication: Gaussian and robust Kronecker product covariance estimation: existence and uniqueness

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q290708)