Gaussian approximation of suprema of empirical processes

DOI10.1214/14-AOS1230zbMATH Open1317.60038arXiv1212.6885MaRDI QIDQ464197FDOQ464197

Authors: Kengo Kato, Victor Chernozhukov, Denis Chetverikov 

Publication date: 17 October 2014

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

Abstract: This paper develops a new direct approach to approximating suprema of general empirical processes by a sequence of suprema of Gaussian processes, without taking the route of approximating whole empirical processes in the sup-norm. We prove an abstract approximation theorem applicable to a wide variety of statistical problems, such as construction of uniform confidence bands for functions. Notably, the bound in the main approximation theorem is nonasymptotic and the theorem allows for functions that index the empirical process to be unbounded and have entropy divergent with the sample size. The proof of the approximation theorem builds on a new coupling inequality for maxima of sums of random vectors, the proof of which depends on an effective use of Stein's method for normal approximation, and some new empirical process techniques. We study applications of this approximation theorem to local and series empirical processes arising in nonparametric estimation via kernel and series methods, where the classes of functions change with the sample size and are non-Donsker. Importantly, our new technique is able to prove the Gaussian approximation for the supremum type statistics under weak regularity conditions, especially concerning the bandwidth and the number of series functions, in those examples.

Full work available at URL: https://arxiv.org/abs/1212.6885

Recommendations

• Gaussian approximation of local empirical processes indexed by functions
• About conditions of Gaussian approximation of kernel estimates for distribution density
• Empirical and multiplier bootstraps for suprema of empirical processes of increasing complexity, and related Gaussian couplings
• On convergence rates of suprema
• Gaussian approximation of the empirical process under random entropy conditions

zbMATH Keywords

Gaussian approximationkernel estimationempirical processesseries estimationcoupling

Mathematics Subject Classification ID

Asymptotic properties of nonparametric inference (62G20) Gaussian processes (60G15) Approximations to statistical distributions (nonasymptotic) (62E17) Functional limit theorems; invariance principles (60F17)

Cites Work

• Weak convergence and empirical processes. With applications to statistics
• Confidence bands in density estimation
• Strong approximation of density estimators from weakly dependent observations by density estimators from independent observations
• Rates of strong uniform consistency for multivariate kernel density estimators. (Vitesse de convergence uniforme presque sûre pour des estimateurs à noyaux de densités multivariées)
• On some global measures of the deviations of density function estimates
• Comparison and anti-concentration bounds for maxima of Gaussian random vectors
• Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
• Uniform in bandwidth consistency of kernel-type function estimators
• Title not available (Why is that?)
• An approximation of partial sums of independent RV'-s, and the sample DF. I
• Concentration inequalities. A nonasymptotic theory of independence
• U-processes: Rates of convergence
• New concentration inequalities in product spaces
• A local maximal inequality under uniform entropy
• Wavelets on the interval and fast wavelet transforms
• Convergence rates and asymptotic normality for series estimators
• Projection estimation in multiple regression with application to functional ANOVA models
• Testing monotonicity of regression.
• Intersection bounds: estimation and inference
• Normal Approximation by Stein’s Method
• On the Error of the Gaussian Approximation for Convolutions
• Uniform Central Limit Theorems
• Gaussian approximation of suprema of empirical processes
• Title not available (Why is that?)
• Anti-concentration and honest, adaptive confidence bands
• Title not available (Why is that?)
• On the Gaussian approximation of convolutions under multidimensional analogues of S. N. Bernstein's inequality conditions
• Invariance principles for sums of Banach space valued random elements and empirical processes
• The Generic Chaining
• Title not available (Why is that?)
• An empirical process approach to the uniform consistency of kernel-type function estimators
• On consistency of kernel density estimators for randomly censored data: Rates holding uniformly over adaptive intervals
• Uniform limit theorems for wavelet density estimators
• On Stein's method for multivariate normal approximation
• Multivariate normal approximation using exchangeable pairs
• Title not available (Why is that?)
• Gaussian approximation of local empirical processes indexed by functions
• Multivariate normal approximation with Stein's method of exchangeable pairs under a general linearity condition
• Moment inequalities for functions of independent random variables
• The Sherrington-Kirkpatrick model
• A user's guide to measure theoretic probability
• Asymptotics for polynomial spline regression under weak conditions.
• Concentration inequalities and confidence bands for needlet density estimators on compact homogeneous manifolds
• Limit theorems for sums of weakly dependent Banach space valued random variables
• Hungarian constructions from the nonasymptotic viewpoint
• Strong approximation for multivariate empirical and related processes, via KMT constructions
• Komlós-Major-Tusnády approximation for the general empirical process and Haar expansions of classes of functions
• Local invariance principles and their application to density estimation
• Functional laws of the iterated logarithm for local empirical processes indexed by sets
• A uniform functional law of the logarithm for the local empirical process.
• Title not available (Why is that?)
• Global uniform risk bounds for wavelet deconvolution estimators
• Multiscale methods for shape constraints in deconvolution: confidence statements for qualitative features
• Revisiting two strong approximation results of Dudley and Philipp
• On the convergence rate of maximal deviation distribution for kernel regression estimates
• Gaussian approximation of the empirical process under random entropy conditions
• A refinement of the KMT inequality for the uniform empirical process
• Stochastic convergence of persistence landscapes and silhouettes
• Rate of convergence in the central limit theorem for empirical processes

Cited In (98)

• A probability approximation framework: Markov process approach
• Notes on the dimension dependence in high-dimensional central limit theorems for hyperrectangles
• Are discoveries spurious? Distributions of maximum spurious correlations and their applications
• About tests of the ``simplifying assumption for conditional copulas
• Space partitioning and regression maxima seeking via a mean-shift-inspired algorithm
• Jackknife multiplier bootstrap: finite sample approximations to the \(U\)-process supremum with applications
• Asymptotic confidence regions for density ridges
• Locally adaptive confidence bands
• Bootstrapping the operator norm in high dimensions: error estimation for covariance matrices and sketching
• Stratified incomplete local simplex tests for curvature of nonparametric multiple regression
• A unified theory of confidence regions and testing for high-dimensional estimating equations
• Robust uniform inference for quantile treatment effects in regression discontinuity designs
• Bootstrap confidence bands for spectral estimation of Lévy densities under high-frequency observations
• A bootstrap method for error estimation in randomized matrix multiplication
• Nonparametric modal regression
• V.N. Sudakov’s Work on Expected Suprema of Gaussian Processes
• Uniformly valid post-regularization confidence regions for many functional parameters in z-estimation framework
• Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
• Nonparametric inference via bootstrapping the debiased estimator
• Multiscale scanning in inverse problems
• Necessary and sufficient conditions for variable selection consistency of the Lasso in high dimensions
• Non-separable models with high-dimensional data
• A massive data framework for M-estimators with cubic-rate
• High-dimensional CLT: improvements, non-uniform extensions and large deviations
• Berry-Esseen bounds for Chernoff-type nonstandard asymptotics in isotonic regression
• Honest confidence sets in nonparametric IV regression and other ill-posed models
• Uniform confidence bands for nonparametric errors-in-variables regression
• Gaussian approximation of maxima of Wiener functionals and its application to high-frequency data
• Monotonicity-constrained nonparametric estimation and inference for first-price auctions
• Unconditional quantile regression with high‐dimensional data
• Title not available (Why is that?)
• Empirical process results for exchangeable arrays
• Testing for monotonicity under endogeneity: an application to the reservation wage function
• Uniform confidence bands in deconvolution with unknown error distribution
• Large Sample Properties of Partitioning-Based Series Estimators
• Approximation to stable law by the Lindeberg principle
• Gaussian approximation of the empirical process under random entropy conditions
• Uniform nonparametric inference for time series
• Volatility coupling
• Hypothesis testing in large-scale functional linear regression
• Nonparametric significance testing in measurement error models
• Bandwidth selection for nonparametric regression with errors-in-variables
• Inference for first-price auctions with Guerre, Perrigne, and Vuong's estimator
• Trinity tests of functions for conditional moment models
• Post-regularization inference for time-varying nonparanormal graphical models
• Kernel meets sieve: post-regularization confidence bands for sparse additive model
• On continuity and strict increase of the CDF for the sup-functional of a Gaussian process with applications to statistics
• Nearly optimal central limit theorem and bootstrap approximations in high dimensions
• Pivotal estimation via square-root lasso in nonparametric regression
• Optimal Gaussian approximation for multiple time series
• Multidimensional multiscale scanning in exponential families: limit theory and statistical consequences
• High-dimensional linear models with many endogenous variables
• On linearization of nonparametric deconvolution estimators for repeated measurements model
• Lasso-driven inference in time and space
• High-dimensional central limit theorems for homogeneous sums
• Empirical and multiplier bootstraps for suprema of empirical processes of increasing complexity, and related Gaussian couplings
• Asymptotic theory for density ridges
• Concentration inequalities for suprema of unbounded empirical processes
• Extremes of locally stationary Gaussian and chi fields on manifolds
• Gaussian approximations for maxima of random vectors under \((2+\iota)\)-th moments
• Quantile regression approach to conditional mode estimation
• Gaussian approximation of suprema of empirical processes
• Some new asymptotic theory for least squares series: pointwise and uniform results
• Proximal Learning for Individualized Treatment Regimes Under Unmeasured Confounding
• Estimation of complier expected shortfall treatment effects with a binary instrumental variable
• Off-Policy Confidence Interval Estimation with Confounded Markov Decision Process
• Estimation and inference for moments of ratios with robustness against large trimming bias
• Rates of the strong uniform consistency with rates for conditional \(U\)-statistics estimators with general kernels on manifolds
• Boundary adaptive local polynomial conditional density estimators
• Large-dimensional central limit theorem with fourth-moment error bounds on convex sets and balls
• Limit distribution theory for smooth \(p\)-Wasserstein distances
• Estimating and inferring the maximum degree of stimulus‐locked time‐varying brain connectivity networks
• Gaussian Approximation and Spatially Dependent Wild Bootstrap for High-Dimensional Spatial Data
• Sharp adaptive and pathwise stable similarity testing for scalar ergodic diffusions
• Nonparametric inference on Lévy measures of compound Poisson-driven Ornstein-Uhlenbeck processes under macroscopic discrete observations
• Confidence bands for survival curves from outcome-dependent stratified samples
• Estimation and inference for policy relevant treatment effects
• Bootstrap Adjustment to Minimum p-Value Method for Predictive Classification
• Bootstrap Inference for Quantile-based Modal Regression
• Comment on “A Tuning-Free Robust and Efficient Approach to High-Dimensional Regression”
• Inference on heterogeneous treatment effects in high‐dimensional dynamic panels under weak dependence
• High-Dimensional Mixed-Frequency IV Regression
• Transformation Models in High Dimensions
• Communication‐efficient low‐dimensional parameter estimation and inference for high‐dimensional Lp$$ {L}^p $$‐quantile regression
• Local regression distribution estimators
• Testing Directed Acyclic Graph via Structural, Supervised and Generative Adversarial Learning
• Value Enhancement of Reinforcement Learning via Efficient and Robust Trust Region Optimization
• The hazard level set
• Estimation and inference of treatment effects with \(L_2\)-boosting in high-dimensional settings
• Inference in nonparametric series estimation with specification searches for the number of series terms
• Bootstrap adjusted predictive classification for identification of subgroups with differential treatment effects under generalized linear models
• Title not available (Why is that?)
• Testing stochastic dominance with many conditioning variables
• Estimation of Conditional Average Treatment Effects With High-Dimensional Data
• Inference on individual treatment effects in nonseparable triangular models
• On semiparametric inference for periodically modulated density functions
• Statistical inference for nonparametric censored regression
• Inference for High-Dimensional Exchangeable Arrays

This page was built for publication: Gaussian approximation of suprema of empirical processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q464197)