Gaussian and bootstrap approximations for high-dimensional U-statistics and their applications

TL;DR

This paper develops Gaussian and bootstrap approximation methods for high-dimensional U-statistics, providing explicit convergence rates and validating their use in ultra-high-dimensional settings where the dimension exceeds the sample size.

Contribution

It introduces a two-step Gaussian approximation procedure applicable without structural assumptions and establishes the validity of multiple bootstrap methods for high-dimensional U-statistics.

Findings

Explicit polynomial decay rate of convergence in high dimensions

Validity of empirical, reweighted, and Gaussian multiplier bootstrap methods

Asymptotic validity when dimension grows exponentially with sample size

Abstract

This paper studies the Gaussian and bootstrap approximations for the probabilities of a non-degenerate U-statistic belonging to the hyperrectangles in $\mathbb{R}^d$ when the dimension $d$ is large. A two-step Gaussian approximation procedure that does not impose structural assumptions on the data distribution is proposed. Subject to mild moment conditions on the kernel, we establish the explicit rate of convergence uniformly in the class of all hyperrectangles in $\mathbb{R}^d$ that decays polynomially in sample size for a high-dimensional scaling limit, where the dimension can be much larger than the sample size. We also provide computable approximation methods for the quantiles of the maxima of centered U-statistics. Specifically, we provide a unified perspective for the empirical bootstrap, the randomly reweighted bootstrap, and the Gaussian multiplier bootstrap with the jackknife estimator of covariance matrix as randomly reweighted quadratic forms and we establish their validity. We show that all three methods are inferentially first-order equivalent for high-dimensional U-statistics in the sense that they achieve the same uniform rate of convergence over all $d$-dimensional hyperrectangles. In particular, they are asymptotically valid when the dimension $d$ can be as large as $O(e^{n^c})$ for some constant $c \in (0,1/7)$. (Full abstract can be found in the paper.)