Concentration inequalities and moment bounds for sample covariance operators
From MaRDI portal
(Redirected from Publication:502859)
Abstract: Let be i.i.d. centered Gaussian random variables in a separable Banach space with covariance operator Sigma:E^{ast}mapsto E, Sigma u = {mathbb E}langle X,u
angle, uin E^{ast}. The sample covariance operator is defined as hat Sigma u := n^{-1}sum_{j=1}^n langle X_j,u
angle X_j, uin E^{ast}. The goal of the paper is to obtain concentration inequalities and expectation bounds for the operator norm of the deviation of the sample covariance operator from the true covariance operator. In particular, it is shown that {mathbb E}|hat Sigma-Sigma|asymp |Sigma|�iggl(sqrt{frac{{�f r}(Sigma)}{n}}�igvee frac{{�f r}(Sigma)}{n}�iggr), where {�f r}(Sigma):=frac{Bigl({mathbb E}|X|Bigr)^2}{|Sigma|}. Moreover, under the assumption that it is proved that, for all with probability at least �egin{align*} Bigl||hatSigma - Sigma|-{mathbb E}|hatSigma - Sigma|Bigr| lesssim |Sigma|�iggl(sqrt{frac{t}{n}}�igvee frac{t}{n}�iggr). end{align*}
Let \(X, X_1, X_2,\ldots, X_n,\ldots\) be i.i.d. (independent and identically distributed) centered random variables in a separable Banach space with covariance operator \(\Sigma\). Also, let \(\hat{\Sigma}\) be the sample covariance operator based on \(X_1,X_2,\ldots,X_n\). Moreover, define \(r(\Sigma)=\frac{(\operatorname{E}\| X\|)^2}{\|\Sigma\|}\). It is shown that \(\operatorname{E}\|\hat{\Sigma}-\Sigma\|\|\Sigma \|\Big(\sqrt{}\frac{r(\Sigma)}{n}\vee\frac{r(\Sigma)}{n}\Big)\). Moreover, under an additional assumption that \(r(\Sigma)\lesssim n\), for some \(C>0\) and for all \(t\geq 1\), the concentration inequality \[ |\|\hat{\Sigma}-\Sigma \|-\operatorname{E}\| \hat{\Sigma-\Sigma \| }|\leq C \| \Sigma \| \Bigg(\sqrt{}\frac{t}{n}\vee \frac{t}{n} \Bigg) \] holds with probability at least \(1-e^{-t}\). Similarly, under the assumption that \(r(\Sigma)\gtrsim n\), the resulting concentration inequality \[ |\| \hat{\Sigma}-\Sigma \|-\operatorname{E}\| \hat{\Sigma}-\Sigma \||\leq C \| \Sigma \|\Bigg(\sqrt{}\frac{r(\Sigma)}{n}\sqrt{}\frac{t}{n}\vee \frac{t}{n}\Bigg) \] holds with the same probability \(1-e^{-t}\).
Recommendations
- Normal approximation and concentration of spectral projectors of sample covariance
- New concentration inequalities for suprema of empirical processes
- Relative perturbation bounds with applications to empirical covariance operators
- Asymptotics and concentration bounds for bilinear forms of spectral projectors of sample covariance
- Concentration inequalities for random tensors
Cited in
(only showing first 100 items - show all)- Bayesian inference for spectral projectors of the covariance matrix
- New asymptotic results in principal component analysis
- Confidence sets for spectral projectors of covariance matrices
- Sub-Gaussian estimators of the mean of a random matrix with heavy-tailed entries
- Hanson-Wright inequality in Banach spaces
- Robust high-dimensional factor models with applications to statistical machine learning
- Approximating \(L_p\) unit balls via random sampling
- Time-uniform, nonparametric, nonasymptotic confidence sequences
- Minimax rates in sparse, high-dimensional change point detection
- Finite impulse response models: a non-asymptotic analysis of the least squares estimator
- Efficient estimation of smooth functionals in Gaussian shift models
- Estimation of smooth functionals in normal models: bias reduction and asymptotic efficiency
- An elementary analysis of ridge regression with random design
- Exact minimax risk for linear least squares, and the lower tail of sample covariance matrices
- An \({\ell_p}\) theory of PCA and spectral clustering
- New challenges in covariance estimation: multiple structures and coarse quantization
- Bootstrapping the operator norm in high dimensions: error estimation for covariance matrices and sketching
- On consistency and sparsity for high-dimensional functional time series with application to autoregressions
- Relative perturbation bounds with applications to empirical covariance operators
- Covariance estimation under one-bit quantization
- On the non-asymptotic concentration of heteroskedastic Wishart-type matrix
- All-in-one robust estimator of the Gaussian mean
- CDPA: common and distinctive pattern analysis between high-dimensional datasets
- Model assisted variable clustering: minimax-optimal recovery and algorithms
- Efficient estimation of linear functionals of principal components
- Uniform Hanson-Wright type concentration inequalities for unbounded entries via the entropy method
- Time-uniform Chernoff bounds via nonnegative supermartingales
- Nonasymptotic upper bounds for the reconstruction error of PCA
- Bootstrapping max statistics in high dimensions: near-parametric rates under weak variance decay and application to functional and multinomial data
- Robust covariance estimation under \(L_4\)-\(L_2\) norm equivalence
- Finite-sample analysis of \(M\)-estimators using self-concordance
- Estimating covariance and precision matrices along subspaces
- MSE bounds for estimators of matrix functions
- Convergence rate of Krasulina estimator
- Robust modifications of U-statistics and applications to covariance estimation problems
- Distributed estimation of principal eigenspaces
- Bootstrap confidence sets for spectral projectors of sample covariance
- The two-to-infinity norm and singular subspace geometry with applications to high-dimensional statistics
- Asymptotically efficient estimation of smooth functionals of covariance operators
- SONIC: social network analysis with influencers and communities
- On convergence rates of adaptive ensemble Kalman inversion for linear ill-posed problems
- Detecting approximate replicate components of a high-dimensional random vector with latent structure
- On cumulative slicing estimation for high dimensional data
- Non-zero constraints in elliptic PDE with random boundary values and applications to hybrid inverse problems
- scientific article; zbMATH DE number 7625199 (Why is no real title available?)
- Dimensionality Reduction, Regularization, and Generalization in Overparameterized Regressions
- Convexification with bounded gap for randomly projected quadratic optimization
- Cross-validation with confidence
- Perturbation bounds for eigenspaces under a relative gap condition
- Normal approximation and concentration of spectral projectors of sample covariance
- The Sup-norm Perturbation of HOSVD and Low Rank Tensor Denoising
- Wald Statistics in high-dimensional PCA
- Deep learning: a statistical viewpoint
- A note on the prediction error of principal component regression in high dimensions
- Learning Gaussian graphical models with latent confounders
- Model Reduction for Nonlinear Systems by Balanced Truncation of State and Gradient Covariance
- Matrix concentration inequalities and free probability
- Nonasymptotic one- and two-sample tests in high dimension with unknown covariance structure
- Contiguity under high-dimensional Gaussianity with applications to covariance testing
- Robustifying Markowitz
- Fast randomized numerical rank estimation for numerically low-rank matrices
- Optimal multiple change-point detection for high-dimensional data
- Dimension-free bounds for sums of independent matrices and simple tensors via the variational principle
- Mean estimation in high dimension
- Dimension-free bounds for sums of dependent matrices and operators with heavy-tailed distributions
- Noise covariance estimation in multi-task high-dimensional linear models
- Composite Index Construction with Expert Opinion
- Large-dimensional central limit theorem with fourth-moment error bounds on convex sets and balls
- Tangent space and dimension estimation with the Wasserstein distance
- Covariance estimation under missing observations and L₄ - L₂ moment equivalence
- Quantitative limit theorems and bootstrap approximations for empirical spectral projectors
- Optimal estimation of Schatten norms of a rectangular matrix
- Improved covariance estimation: optimal robustness and sub-Gaussian guarantees under heavy tails
- Comparing regularisation paths of (conjugate) gradient estimators in ridge regression
- Pseudo-labeling for kernel ridge regression under covariate shift
- Covariance estimation: optimal dimension-free guarantees for adversarial corruption and heavy tails
- Matrix Freedman inequality for sub-Weibull martingales
- Estimation and inference for CP tensor factor models
- Benign overfitting in time-series linear models with over-parameterization
- Sharp concentration of simple random tensors
- Davis-Kahan theorem under a moderate gap condition
- Hub Detection in Gaussian Graphical Models
- Upper bounds of spiked covariance matrices under differentially private constrains
- On the dimension-free concentration of simple tensors via matrix deviation
- On the estimation of Gaussian moment tensors
- Mode-wise principal subspace pursuit and matrix spiked covariance model
- Concentration and consistency of sample covariance matrix functionals in sub-Gaussian models
- Covariance operator estimation via adaptive thresholding
- Precision and Cholesky factor estimation for Gaussian processes
- Estimators for multivariate allometric regression model
- Improved rates of bootstrap approximation for the operator norm: a coordinate-free approach
- Statistically optimal robust mean and covariance estimation for anisotropic gaussians
- On the multiway principal component analysis
- Stable rank and intrinsic dimension of real and complex matrices
- Generalisation bounds for kernel PCA through PAC-Bayes learning
- A parameter-free two-bit covariance estimator with improved operator norm error rate
- Large-dimensional independent component analysis: statistical optimality and computational tractability
- Covariance operator estimation: sparsity, lengthscale, and ensemble Kalman filters
- Spectral norm of exponentially weighted moving sample covariance matrix and its application to sequential sparse signal detection
- The impact of contamination and correlated design on the Lasso: an average case analysis
This page was built for publication: Concentration inequalities and moment bounds for sample covariance operators
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q502859)
