High-dimensional CLT: Improvements, Non-uniform Extensions and Large Deviations

TL;DR

This paper advances high-dimensional CLTs by improving convergence rates, extending to non-uniform cases, and applying results to post-selection inference and empirical processes.

Contribution

It improves the convergence rate bounds for high-dimensional CLTs and introduces non-uniform variants based on large deviation principles.

Findings

Improved the rate of convergence from $ ext{log}^7 p = o(n)$ to $ ext{log}^4 p = o(n)$.

Established non-uniform CLTs for high-dimensional vectors using large deviation techniques.

Applied the theoretical results to post-selection inference and empirical process analysis.

Abstract

Central limit theorems (CLTs) for high-dimensional random vectors with dimension possibly growing with the sample size have received a lot of attention in the recent times. Chernozhukov et al. (2017) proved a Berry--Esseen type result for high-dimensional averages for the class of hyperrectangles and they proved that the rate of convergence can be upper bounded by $n^{-1/6}$ upto a polynomial factor of $\log p$ (where $n$ represents the sample size and $p$ denotes the dimension). Convergence to zero of the bound requires $\log^7p = o(n)$. We improve upon their result which only requires $\log^4p = o(n)$ (in the best case). This improvement is made possible by a sharper dimension-free anti-concentration inequality for Gaussian process on a compact metric space. In addition, we prove two non-uniform variants of the high-dimensional CLT based on the large deviation and non-uniform CLT results for random variables in a Banach space by Bentkus, Ra{\v c}kauskas, and Paulauskas. We apply our results in the context of post-selection inference in linear regression and of empirical processes.