Minimax sparse principal subspace estimation in high dimensions

DOI10.1214/13-AOS1151zbMath1288.62103arXiv1211.0373MaRDI QIDQ2443207

Vincent Q. Vu, Jing Lei

Publication date: 4 April 2014

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

Full work available at URL: https://arxiv.org/abs/1211.0373

zbMATH Keywords

random matricessparsityhigh-dimensional statisticsminimax boundsprincipal components analysis

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Estimation in multivariate analysis (62H12) Minimax procedures in statistical decision theory (62C20) Eigenvalues, singular values, and eigenvectors (15A18)

Related Items (55)

Testing equivalence of clustering ⋮ Wald Statistics in high-dimensional PCA ⋮ High-Dimensional Vector Autoregressive Time Series Modeling via Tensor Decomposition ⋮ Overlapping community detection in networks via sparse spectral decomposition ⋮ A Schatten-\(q\) low-rank matrix perturbation analysis via perturbation projection error bound ⋮ High-resolution signal recovery via generalized sampling and functional principal component analysis ⋮ Sparse Functional Principal Component Analysis in High Dimensions ⋮ Alternating maximization: unifying framework for 8 sparse PCA formulations and efficient parallel codes ⋮ Optimal sparse eigenspace and low-rank density matrix estimation for quantum systems ⋮ Sparse PCA-based on high-dimensional Itô processes with measurement errors ⋮ Lower bounds for invariant statistical models with applications to principal component analysis ⋮ Recent developments in high dimensional covariance estimation and its related issues, a review ⋮ Efficient estimation of linear functionals of principal components ⋮ Large covariance estimation through elliptical factor models ⋮ Minimax estimation in sparse canonical correlation analysis ⋮ Nearly optimal stochastic approximation for online principal subspace estimation ⋮ [HDDA sparse subspace constrained partial least squares] ⋮ Distributed Estimation for Principal Component Analysis: An Enlarged Eigenspace Analysis ⋮ Stochastic Gauss-Newton algorithms for online PCA ⋮ Van Trees inequality, group equivariance, and estimation of principal subspaces ⋮ Bayesian sparse spiked covariance model with a continuous matrix shrinkage prior ⋮ Unnamed Item ⋮ Optimal discriminant analysis in high-dimensional latent factor models ⋮ Sparse principal component analysis for high‐dimensional stationary time series ⋮ Nonasymptotic upper bounds for the reconstruction error of PCA ⋮ Dynamic Principal Component Analysis in High Dimensions ⋮ Euclidean Representation of Low-Rank Matrices and Its Geometric Properties ⋮ On the optimality of the Oja's algorithm for online PCA ⋮ PCA Sparsified ⋮ Long random matrices and tensor unfolding ⋮ Envelopes and principal component regression ⋮ Robust covariance estimation for approximate factor models ⋮ Near-optimal stochastic approximation for online principal component estimation ⋮ Posterior contraction rates of the phylogenetic Indian buffet processes ⋮ Slice inverse regression with score functions ⋮ An $\ell_{\infty}$ Eigenvector Perturbation Bound and Its Application to Robust Covariance Estimation ⋮ Subspace perspective on canonical correlation analysis: dimension reduction and minimax rates ⋮ Bayesian variable selection for globally sparse probabilistic PCA ⋮ ECA: High-Dimensional Elliptical Component Analysis in Non-Gaussian Distributions ⋮ Distributed estimation of principal eigenspaces ⋮ Linearity identification for general partial linear single-index models ⋮ Recovery of simultaneous low rank and two-way sparse coefficient matrices, a nonconvex approach ⋮ Estimating structured high-dimensional covariance and precision matrices: optimal rates and adaptive estimation ⋮ High-dimensional index volatility models via Stein's identity ⋮ Principal component analysis in the local differential privacy model ⋮ Robust covariance and scatter matrix estimation under Huber's contamination model ⋮ Sparse principal component analysis with missing observations ⋮ Spectral thresholding for the estimation of Markov chain transition operators ⋮ Squared-norm empirical processes ⋮ Consistency of spectral clustering in stochastic block models ⋮ Sparsistency and agnostic inference in sparse PCA ⋮ Optimal estimation and rank detection for sparse spiked covariance matrices ⋮ Rate-optimal posterior contraction for sparse PCA ⋮ A Unifying Tutorial on Approximate Message Passing ⋮ Do semidefinite relaxations solve sparse PCA up to the information limit?

Uses Software

• PMA

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Sparse principal component analysis and iterative thresholding
• Minimax bounds for sparse PCA with noisy high-dimensional data
• Empirical processes with a bounded \(\psi_1\) diameter
• Coordinate-independent sparse sufficient dimension reduction and variable selection
• High-dimensional analysis of semidefinite relaxations for sparse principal components
• Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
• Sparse principal component analysis via regularized low rank matrix approximation
• Finite sample approximation results for principal component analysis: A matrix perturbation approach
• The composite absolute penalties family for grouped and hierarchical variable selection
• Minimax risk over \(l_ p\)-balls for \(l_ q\)-error
• Statistics on special manifolds
• Weak convergence and empirical processes. With applications to statistics
• Consistency of sparse PCA in high dimension, low sample size contexts
• Sparse Principal Component Analysis with Missing Observations
• Generalized power method for sparse principal component analysis
• The Geometry of Algorithms with Orthogonality Constraints
• On Consistency and Sparsity for Principal Components Analysis in High Dimensions
• Minimax Rates of Estimation for High-Dimensional Linear Regression Over $\ell_q$-Balls
• Model Selection and Estimation in Regression with Grouped Variables
• A Direct Formulation for Sparse PCA Using Semidefinite Programming
• A unified framework for high-dimensional analysis of \(M\)-estimators with decomposable regularizers

This page was built for publication: Minimax sparse principal subspace estimation in high dimensions