Concentration inequalities for statistical inference

DOI10.4208/CMR.2020-0041zbMATH Open1488.60028arXiv2011.02258OpenAlexW3128339971MaRDI QIDQ3380883FDOQ3380883

Song Xi Chen, Huiming Zhang

Publication date: 29 September 2021

Published in: Communications in Mathematical Research (Search for Journal in Brave)

Abstract: This paper gives a review of concentration inequalities which are widely employed in non-asymptotical analyses of mathematical statistics in a wide range of settings, from distribution-free to distribution-dependent, from sub-Gaussian to sub-exponential, sub-Gamma, and sub-Weibull random variables, and from the mean to the maximum concentration. This review provides results in these settings with some fresh new results. Given the increasing popularity of high-dimensional data and inference, results in the context of high-dimensional linear and Poisson regressions are also provided. We aim to illustrate the concentration inequalities with known constants and to improve existing bounds with sharper constants.

Full work available at URL: https://arxiv.org/abs/2011.02258

Recommendations

• Advanced Lectures on Machine Learning
• Concentration inequalities. A nonasymptotic theory of independence
• Preface
• Concentration inequalities, counting processes and adaptive statistics
• Some applications of concentration inequalities to statistics

zbMATH Keywords

random matricesheavy-tailed distributionsfinite-sample theoryconstants-specified inequalitieshigh-dimensional estimation and testingsub-Weibull random variables

Mathematics Subject Classification ID

Inequalities; stochastic orderings (60E15) Approximations to statistical distributions (nonasymptotic) (62E17) Sums of independent random variables; random walks (60G50)

Cites Work

• Weak convergence and empirical processes. With applications to statistics
• Asymptotic Statistics
• Title not available (Why is that?)
• Statistics for high-dimensional data. Methods, theory and applications.
• Simultaneous analysis of Lasso and Dantzig selector
• The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
• Title not available (Why is that?)
• Kullback-Leibler aggregation and misspecified generalized linear models
• Univariate Discrete Distributions
• Lasso guarantees for \(\beta \)-mixing heavy-tailed time series
• Mathematical Foundations of Infinite-Dimensional Statistical Models
• Title not available (Why is that?)
• High-Dimensional Statistics
• High-Dimensional Probability
• Title not available (Why is that?)
• Probability Inequalities for Sums of Bounded Random Variables
• DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
• Hanson-Wright inequality and sub-Gaussian concentration
• The tight constant in the Dvoretzky-Kiefer-Wolfowitz inequality
• Title not available (Why is that?)
• Oracle inequalities in empirical risk minimization and sparse recovery problems. École d'Été de Probabilités de Saint-Flour XXXVIII-2008.
• A tail inequality for quadratic forms of subgaussian random vectors
• Sharper bounds for Gaussian and empirical processes
• A two-sample test for high-dimensional data with applications to gene-set testing
• Asymptotic Minimax Character of the Sample Distribution Function and of the Classical Multinomial Estimator
• Title not available (Why is that?)
• Oracle inequalities for the lasso in the Cox model
• Title not available (Why is that?)
• Non-asymptotic oracle inequalities for the high-dimensional cox regression via lasso
• Learning Bounds for Kernel Regression Using Effective Data Dimensionality
• Limit of the smallest eigenvalue of a large dimensional sample covariance matrix
• Nonparametric statistics for stochastic processes. Estimation and prediction.
• Rate minimaxity of the Lasso and Dantzig selector for the \(l_{q}\) loss in \(l_{r}\) balls
• Title not available (Why is that?)
• Title not available (Why is that?)
• Some nonasymptotic results on resampling in high dimension. I: Confidence regions
• \(\ell _{1}\)-regularized linear regression: persistence and oracle inequalities
• Honest variable selection in linear and logistic regression models via \(\ell _{1}\) and \(\ell _{1}+\ell _{2}\) penalization
• On the sample covariance matrix estimator of reduced effective rank population matrices, with applications to fPCA
• Probability Inequalities for the Sum of Independent Random Variables
• A Bound on Tail Probabilities for Quadratic Forms in Independent Random Variables
• A tail inequality for suprema of unbounded empirical processes with applications to Markov chains
• Log-concave probability and its applications
• An Introduction to Heavy-Tailed and Subexponential Distributions
• On learning with integral operators
• High dimensional robust M-estimation: asymptotic variance via approximate message passing
• Title not available (Why is that?)
• Large Sample Covariance Matrices and High-Dimensional Data Analysis
• Tests for High-Dimensional Regression Coefficients With Factorial Designs
• Sums and Gaussian vectors
• On robust regression with high-dimensional predictors
• Title not available (Why is that?)
• Random matrices: sharp concentration of eigenvalues
• Log-concavity and strong log-concavity: a review
• Concentration and Moment Inequalities for Polynomials of Independent Random Variables
• Upper (lower) bounds on the mean of the maximum (minimum) of a number of random variables
• Concentration inequalities for order statistics
• Title not available (Why is that?)
• Probability
• Linear Hypothesis Testing in Dense High-Dimensional Linear Models
• Tail bounds for the supremums of empirical processes over unbounded classes of functions
• Exponential integrability of sub-Gaussian vectors
• Asymptotics for high dimensional regression \(M\)-estimates: fixed design results
• Notes on discrete compound Poisson model with applications to risk theory
• The Bennett-Orlicz norm
• Statistical Foundations of Data Science
• Square‐Root LASSO for High‐Dimensional Sparse Linear Systems with Weakly Dependent Errors
• Asymptotic theory for differentially private generalized \(\beta\)-models with parameters increasing
• Weighted Lasso estimates for sparse logistic regression: non-asymptotic properties with measurement errors
• Elastic-net Regularized High-dimensional Negative Binomial Regression: Consistency and Weak Signal Detection
• L1 least squares for sparse high-dimensional LDA
• Title not available (Why is that?)

Cited In (17)

• Some applications of concentration inequalities to statistics
• Elastic-net Regularized High-dimensional Negative Binomial Regression: Consistency and Weak Signal Detection
• On the sub-Gaussianity of the missing mass
• Asymptotic in the ordered networks with a noisy degree sequence
• Asymptotic theory in network models with covariates and a growing number of node parameters
• Title not available (Why is that?)
• Concentration inequalities of MLE and robust MLE
• Oracle inequalities for weighted group Lasso in high-dimensional Poisson regression model
• Asymptotics in the Thurstone model with an increasing items
• Title not available (Why is that?)
• Non-asymptotic sub-Gaussian error bounds for hypothesis testing
• Optimal distribution-free concentration for the log-likelihood function of Bernoulli variables
• Concentration inequalities and laws of large numbers under epistemic and regular irrelevance
• Asymptotics in the Thurstone model with an increasing items
• Concentration inequalities for upper probabilities
• Asymptotic normality and confidence region for Catoni's Z estimator
• Concentration inequality of maximum likelihood estimator

This page was built for publication: Concentration inequalities for statistical inference

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3380883)