Optimal shrinkage of eigenvalues in the spiked covariance model

DOI10.1214/17-AOS1601zbMath1403.62099arXiv1311.0851OpenAlexW2964044082WikidataQ91786599 ScholiaQ91786599MaRDI QIDQ2413608

Iain M. Johnstone, Matan Gavish, David L. Donoho

Publication date: 14 September 2018

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

Full work available at URL: https://arxiv.org/abs/1311.0851

zbMATH Keywords

entropy lossFréchet distancecovariance estimationspiked covarianceStein lossoptimal shrinkagehigh-dimensional asymptoticsdivergence lossBhattacharya/Matusita affinitycondition number loss

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Eigenvalues, singular values, and eigenvectors (15A18) Random matrices (algebraic aspects) (15B52)

Related Items

Heteroskedastic PCA: algorithm, optimality, and applications, Eigenvalues of stochastic blockmodel graphs and random graphs with low-rank edge probability matrices, Optimally Weighted PCA for High-Dimensional Heteroscedastic Data, Statistical inference for principal components of spiked covariance matrices, Log-determinant divergences revisited: alpha-beta and gamma log-det divergences, Quadratic shrinkage for large covariance matrices, Targeted principal components regression, Multilevel Sequential Monte Carlo with Dimension-Independent Likelihood-Informed Proposals, Disordered systems insights on computational hardness, Unnamed Item, Unnamed Item, The Dispersion Bias, Multidimensional scaling of noisy high dimensional data, Optimal prediction in the linearly transformed spiked model, Optimal singular value shrinkage for operator norm loss: extending to non-square matrices, Free energy subadditivity for symmetric random Hamiltonians, Perturbative construction of mean-field equations in extensive-rank matrix factorization and denoising, Matrix means and a novel high-dimensional shrinkage phenomenon, Regularized matrix data clustering and its application to image analysis, CORRELATION MATRIX OF EQUI-CORRELATED NORMAL POPULATION: FLUCTUATION OF THE LARGEST EIGENVALUE, SCALING OF THE BULK EIGENVALUES, AND STOCK MARKET, Consistent estimation of high-dimensional factor models when the factor number is over-estimated, Estimation of the Number of Spiked Eigenvalues in a Covariance Matrix by Bulk Eigenvalue Matching Analysis, Linear Hypothesis Testing in Linear Models With High-Dimensional Responses, Unnamed Item, STATISTICAL INFERENCE WITH F-STATISTICS WHEN FITTING SIMPLE MODELS TO HIGH-DIMENSIONAL DATA, An Empirical Bayes Approach to Shrinkage Estimation on the Manifold of Symmetric Positive-Definite Matrices, Euclidean Representation of Low-Rank Matrices and Its Geometric Properties, Long random matrices and tensor unfolding, A CLT for the LSS of large-dimensional sample covariance matrices with diverging spikes, Detecting approximate replicate components of a high-dimensional random vector with latent structure, \(e\)PCA: high dimensional exponential family PCA, Rapid evaluation of the spectral signal detection threshold and Stieltjes transform, Tail sums of Wishart and Gaussian eigenvalues beyond the bulk edge, Shrinkage estimation of large covariance matrices: keep it simple, statistician?, Shrinkage priors for single-spiked covariance models, A note on identifiability conditions in confirmatory factor analysis, Think globally, fit locally under the manifold setup: asymptotic analysis of locally linear embedding, Optimal estimation of a large-dimensional covariance matrix under Stein's loss, Structural Variability from Noisy Tomographic Projections, Robust high-dimensional factor models with applications to statistical machine learning, Spiked separable covariance matrices and principal components, High-dimensional linear models: a random matrix perspective, Geodesically Parameterized Covariance Estimation, Random matrix theory and its applications, Design-free estimation of integrated covariance matrices for high-frequency data, Method of moments for 3D single particleab initiomodeling with non-uniform distribution of viewing angles, Latent common manifold learning with alternating diffusion: analysis and applications, Precise statistical analysis of classification accuracies for adversarial training, Spiked sample covariance matrices with possibly multiple bulk components, Graph-Based Regularization for Regression Problems with Alignment and Highly Correlated Designs

Uses Software

• DLMF

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A well-conditioned estimator for large-dimensional covariance matrices
• Nonlinear shrinkage estimation of large-dimensional covariance matrices
• High dimensional covariance matrix estimation using a factor model
• Reconstruction of a low-rank matrix in the presence of Gaussian noise
• Fluctuations of the extreme eigenvalues of finite rank deformations of random matrices
• Minimax risk of matrix denoising by singular value thresholding
• The eigenvalues and eigenvectors of finite, low rank perturbations of large random matrices
• The distance between two random vectors wigh given dispersion matrices
• Eigenvectors of some large sample covariance matrix ensembles
• Central limit theorems for eigenvalues in a spiked population model
• Estimating the covariance matrix and the generalized variance under a symmetric loss
• On estimation of a matrix of normal means with unknown covariance matrix
• Statistics on the manifold of multivariate normal distributions: theory and application to diffusion tensor MRI processing
• Non-Euclidean statistics for covariance matrices, with applications to diffusion tensor imaging
• Operator norm consistent estimation of large-dimensional sparse covariance matrices
• Spectrum estimation for large dimensional covariance matrices using random matrix theory
• Spectral analysis of large dimensional random matrices
• Estimation of a covariance matrix under Stein's loss
• Lectures on the theory of estimation of many parameters
• Improved estimation of a covariance matrix under quadratic loss
• An identity for the Wishart distribution with applications
• A limit theorem for the norm of random matrices
• Estimation of the inverse covariance matrix: Random mixtures of the inverse Wishart matrix and the identity
• Empirical Bayes estimation of the multivariate normal covariance matrix
• The Frechet distance between multivariate normal distributions
• Estimating covariance matrices
• Multivariate empirical Bayes and estimation of covariance matrices
• Minimax risk over \(l_ p\)-balls for \(l_ q\)-error
• Estimation of a covariance matrix using the reference prior
• On the distribution of the largest eigenvalue in principal components analysis
• On orthogonal and symplectic matrix ensembles
• Nonparametric empirical Bayes and compound decision approaches to estimation of a high-dimensional vector of normal means
• Estimation of the multivariate normal precision and covariance matrices in a star-shape model
• Eigenvalues of large sample covariance matrices of spiked population models
• On the notion of affinity of several distributions and some of its applications
• Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices
• The Optimal Hard Threshold for Singular Values is <inline-formula> <tex-math notation="TeX">\(4/\sqrt {3}\) </tex-math></inline-formula>
• Shrinkage Estimators for Covariance Matrices
• Improved minimax estimation of a normal precision matrix
• Non-Parametric Detection of the Number of Signals: Hypothesis Testing and Random Matrix Theory
• Shrinkage Algorithms for MMSE Covariance Estimation
• Estimating the normal dispersion matrix and the precision matrix from a decision-theoretic point of view: a review
• Improved Minimax Estimators of Normal Convariance and Precision Matrices
• Optimal Shrinkage of Singular Values
• On the Bias of Functions of Characteristic Roots of a Random Matrix
• Covariance matrix selection and estimation via penalised normal likelihood
• DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
• On Certain Characteristics of the Distribution of the Latent Roots of a Symmetric Random Matrix Under General Conditions
• Condition-Number-Regularized Covariance Estimation
• Normal Multivariate Analysis and the Orthogonal Group