Dual Induction CLT for High-dimensional m-dependent Data

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Bong, Heejong, Kuchibhotla, Arun Kumar, Rinaldo, Alessandro
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911128321261568
author Bong, Heejong
Kuchibhotla, Arun Kumar
Rinaldo, Alessandro
author_facet Bong, Heejong
Kuchibhotla, Arun Kumar
Rinaldo, Alessandro
contents We derive novel and sharp high-dimensional Berry--Esseen bounds for the sum of $m$-dependent random vectors over the class of hyper-rectangles exhibiting only a poly-logarithmic dependence in the dimension. Our results hold under minimal assumptions, such as non-degenerate covariances and finite third moments, and exhibit an optimal sample complexity of order $m^{(q-1)/(q-2)}/\sqrt{n}$. Aside from logarithmic terms, the resulting rates match the optimal rates established in the univariate case. When specialized to the sums of independent non-degenerate random vectors, our results produce sharp and, in some cases, optimal rates under the weakest possible conditions. We develop a novel inductive relationship between anti-concentration inequalities and Berry--Esseen bounds inspired by the classical Lindeberg swapping method and the concentration inequality approach for dependent data that may be of independent interest.
format Preprint
id arxiv_https___arxiv_org_abs_2306_14299
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Dual Induction CLT for High-dimensional m-dependent Data
Bong, Heejong
Kuchibhotla, Arun Kumar
Rinaldo, Alessandro
Probability
Statistics Theory
60B12, 60F05
We derive novel and sharp high-dimensional Berry--Esseen bounds for the sum of $m$-dependent random vectors over the class of hyper-rectangles exhibiting only a poly-logarithmic dependence in the dimension. Our results hold under minimal assumptions, such as non-degenerate covariances and finite third moments, and exhibit an optimal sample complexity of order $m^{(q-1)/(q-2)}/\sqrt{n}$. Aside from logarithmic terms, the resulting rates match the optimal rates established in the univariate case. When specialized to the sums of independent non-degenerate random vectors, our results produce sharp and, in some cases, optimal rates under the weakest possible conditions. We develop a novel inductive relationship between anti-concentration inequalities and Berry--Esseen bounds inspired by the classical Lindeberg swapping method and the concentration inequality approach for dependent data that may be of independent interest.
title Dual Induction CLT for High-dimensional m-dependent Data
topic Probability
Statistics Theory
60B12, 60F05
url https://arxiv.org/abs/2306.14299