From p-Wasserstein bounds to moderate deviations

DOI10.1214/23-EJP976MaRDI QIDQ6165989zbMATH OpenOpenAlexFDO

Authors Xiao Fang, Yuta Koike

Publication date 2 August 2023

Published in Electronic Journal of Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2205.13307

zbMATH Keywords

\(p\)-Wasserstein distancecentral limit theoremCramér-type moderate deviationsmultivariate normal approximationStein's method

Mathematics Subject Classification ID

Central limit and other weak theorems (60F05) Large deviations (60F10) Approximations to statistical distributions (nonasymptotic) (62E17)

Abstract: We use a new method via $p$-Wasserstein bounds to prove Cram'er-type moderate deviations in (multivariate) normal approximations. In the classical setting that $W$ is a standardized sum of $n$ independent and identically distributed (i.i.d.) random variables with sub-exponential tails, our method recovers the optimal range of $0leqx=o(n^{1/6})$ and the near optimal error rate $O(1)(1+x)(logn+x^2)/sqrtn$ for $P(W>x)/(1-Phi(x))o1$, where $Phi$ is the standard normal distribution function. Our method also works for dependent random variables (vectors) and we give applications to the combinatorial central limit theorem, Wiener chaos, homogeneous sums and local dependence. The key step of our method is to show that the $p$-Wasserstein distance between the distribution of the random variable (vector) of interest and a normal distribution grows like $O(p^{a}lphaDelta)$, $1leqpleq{p}_0$, for some constants $alpha,Delta$ and $p_0$. In the above i.i.d. setting, $alpha=1,Delta=1/sqrtn,p_0=n^{1/3}$. For this purpose, we obtain general $p$-Wasserstein bounds in (multivariate) normal approximations using Stein's method.

Recommendations

• Stein's method for normal approximation in Wasserstein distances with application to the multivariate central limit theorem
• From Stein identities to moderate deviations
• Wasserstein-2 bounds in normal approximation under local dependence
• On the Wasserstein distance for a martingale central limit theorem
• New Berry-Esseen and Wasserstein bounds in the CLT for non-randomly centered random sums by probabilistic methods

Cites work

• A central limit theorem for decomposable random variables with applications to random graphs
• A central limit theorem for generalized multilinear forms
• A high-dimensional CLT in \(\mathcal {W}_2\) distance with near optimal convergence rate
• An approximation of partial sums of independent RV'-s, and the sample DF. I
• Berry-Esseen bounds for generalized U-statistics
• Concentration inequalities for non-Lipschitz functions with bounded derivatives of higher order
• Concentration inequalities. A nonasymptotic theory of independence
• Cramér-type moderate deviation for quadratic forms with a fast rate
• Cumulants on Wiener chaos: moderate deviations and the fourth moment theorem
• Entropy and random vectors
• Entropy and the fourth moment phenomenon
• Estimates of moments and tails of Gaussian chaoses
• From Stein identities to moderate deviations
• Gaussian Hilbert Spaces
• Generalization of an inequality by Talagrand and links with the logarithmic Sobolev inequality
• High-dimensional central limit theorems for homogeneous sums
• High-dimensional probability. An introduction with applications in data science
• scientific article; zbMATH DE number 3174854 (Why is no real title available?)
• scientific article; zbMATH DE number 53996 (Why is no real title available?)
• scientific article; zbMATH DE number 3504209 (Why is no real title available?)
• scientific article; zbMATH DE number 558238 (Why is no real title available?)
• scientific article; zbMATH DE number 3438144 (Why is no real title available?)
• scientific article; zbMATH DE number 6125590 (Why is no real title available?)
• scientific article; zbMATH DE number 7662451 (Why is no real title available?)
• Inference for time-varying lead-lag relationships from ultra-high-frequency data
• Invariance principles for homogeneous sums: universality of Gaussian Wiener chaos
• Inversion of conditional expectations
• Moments of the Gaussian Chaos
• Moving beyond sub-Gaussianity in high-dimensional statistics: applications in covariance estimation and linear regression
• Multi-dimensional integral limit theorems for large deviations
• Multivariate normal approximation with Stein's method of exchangeable pairs under a general linearity condition
• New error bounds in multivariate normal approximations via exchangeable pairs with applications to Wishart matrices and fourth moment theorems
• Normal approximation under local dependence.
• Normal approximations with Malliavin calculus. From Stein's method to universality
• On large deviations for combinatorial sums
• On normal approximations of distributions in terms of dependency graphs
• On the error bound in a combinatorial central limit theorem
• Probability with Martingales
• Quantile coupling inequalities and their applications
• Quantitative de Jong theorems in any dimension
• Rates of convergence for multivariate normal approximation with applications to dense graphs and doubly indexed permutation statistics
• Stein's method for concentration inequalities
• Stein's method for normal approximation in Wasserstein distances with application to the multivariate central limit theorem
• Stein's method, logarithmic Sobolev and transport inequalities
• Suprema of canonical Weibull processes
• The central limit theorem for dependent random variables

Cited in

(6)

• The moderate deviation principle for minimizers of convex processes
• Wasserstein-2 bounds in normal approximation under local dependence
• Wasserstein-p bounds in the central limit theorem under local dependence
• Improved rates of convergence for the multivariate central limit theorem in Wasserstein distance
• Moderate deviations for functionals over infinitely many Rademacher random variables
• High-dimensional bootstrap and asymptotic expansion

This page was built for publication: From \(p\)-Wasserstein bounds to moderate deviations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6165989)