Sharp High-dimensional Central Limit Theorems for Log-concave Distributions

MaRDI QIDQ6406438zbMATH OpenFDO

Authors Xiao Fang, Yuta Koike

Publication date 29 July 2022

Mathematics Subject Classification ID

Central limit and other weak theorems (60F05) Diffusion processes (60J60) Approximations to statistical distributions (nonasymptotic) (62E17)

Abstract: Let $X_1,dots,X_n$ be i.i.d. log-concave random vectors in $mathbb{R}^d$ with mean 0 and covariance matrix $Sigma$. We study the problem of quantifying the normal approximation error for $W=n^{-1/2}su{m}_{i=1}^n{X}_i$ with explicit dependence on the dimension $d$. Specifically, without any restriction on $Sigma$, we show that the approximation error over rectangles in $mathbb{R}^d$ is bounded by $C(lo{g}^{13}(dn)/n{)}^{1/2}$ for some universal constant $C$. Moreover, if the Kannan-Lov'asz-Simonovits (KLS) spectral gap conjecture is true, this bound can be improved to $C(lo{g}^3(dn)/n{)}^{1/2}$. This improved bound is optimal in terms of both $n$ and $d$ in the regime $logn=O(logd)$. We also give $p$-Wasserstein bounds with all $pgeq2$ and a Cram'er type moderate deviation result for this normal approximation error, and they are all optimal under the KLS conjecture. To prove these bounds, we develop a new Gaussian coupling inequality that gives almost dimension-free bounds for projected versions of $p$-Wasserstein distance for every $pgeq2$. We prove this coupling inequality by combining Stein's method and Eldan's stochastic localization procedure.

This page was built for publication: Sharp High-dimensional Central Limit Theorems for Log-concave Distributions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6406438)