Robust Shrinkage Estimation of High-Dimensional Covariance Matrices

DOI10.1109/TSP.2011.2138698zbMATH Open1391.62088arXiv1009.5331MaRDI QIDQ4573154FDOQ4573154

Authors: Yilun Chen, Ami Wiesel, Alfred O. III Hero 

Publication date: 18 July 2018

Published in: IEEE Transactions on Signal Processing (Search for Journal in Brave)

Abstract: We address high dimensional covariance estimation for elliptical distributed samples, which are also known as spherically invariant random vectors (SIRV) or compound-Gaussian processes. Specifically we consider shrinkage methods that are suitable for high dimensional problems with a small number of samples (large $p$ small $n$). We start from a classical robust covariance estimator [Tyler(1987)], which is distribution-free within the family of elliptical distribution but inapplicable when $n<p$. Using a shrinkage coefficient, we regularize Tyler's fixed point iterations. We prove that, for all $n$ and $p$, the proposed fixed point iterations converge to a unique limit regardless of the initial condition. Next, we propose a simple, closed-form and data dependent choice for the shrinkage coefficient, which is based on a minimum mean squared error framework. Simulations demonstrate that the proposed method achieves low estimation error and is robust to heavy-tailed samples. Finally, as a real world application we demonstrate the performance of the proposed technique in the context of activity/intrusion detection using a wireless sensor network.

Full work available at URL: https://arxiv.org/abs/1009.5331

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Ridge regression; shrinkage estimators (Lasso) (62J07) Robustness and adaptive procedures (parametric inference) (62F35)

Cited In (25)

• Relaxing the Gaussian assumption in shrinkage and SURE in high dimension
• An adaptive covariance parameterization technique for the ensemble Gaussian mixture filter
• Large dimensional analysis and optimization of robust shrinkage covariance matrix estimators
• High-dimensional robust precision matrix estimation: cellwise corruption under \(\epsilon \)-contamination
• Gaussian and robust Kronecker product covariance estimation: existence and uniqueness
• Kronecker-structured covariance models for multiway data
• Multivariate outlier detection based on a robust Mahalanobis distance with shrinkage estimators
• A review of Tyler's shape matrix and its extensions
• Title not available (Why is that?)
• Second order statistics of robust estimators of scatter. Application to GLRT detection for elliptical signals
• Tyler's and Maronna's M-estimators: non-asymptotic concentration results
• Conic Geometric Optimization on the Manifold of Positive Definite Matrices
• Marčenko-Pastur law for Tyler's M-estimator
• Weighted Elastic Net Penalized Mean-Variance Portfolio Design and Computation
• The minimum weighted covariance determinant estimator for high-dimensional data
• An overview of heavy-tail extensions of multivariate Gaussian distribution and their relations
• Optimal estimation of a large-dimensional covariance matrix under Stein's loss
• Robust sparse covariance estimation by thresholding Tyler's M-estimator
• Adaptive control and signal processing literature survey (No. 27)
• Shrinkage Estimators for Covariance Matrices
• Quadratic shrinkage for large covariance matrices
• Kernel generalized canonical correlation analysis
• Principal regression for high dimensional covariance matrices
• Multivariate Myriad Filters Based on Parameter Estimation of Student-$t$ Distributions
• Improving portfolios global performance using a cleaned and robust covariance matrix estimate

This page was built for publication: Robust Shrinkage Estimation of High-Dimensional Covariance Matrices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4573154)