Bounding distributional errors via density ratios

DOI10.3150/20-BEJ1256MaRDI QIDQ2040039zbMATH OpenOpenAlexFDO

Authors Lutz Dümbgen, Richard Samworth, Jon A. Wellner

Publication date 9 July 2021

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1905.03009

zbMATH Keywords

binomial distributionhypergeometric distributionPoisson approximationrelative errorstotal variation distance

Mathematics Subject Classification ID

Convergence of probability measures (60B10) Approximations to statistical distributions (nonasymptotic) (62E17)

Abstract: We present some new and explicit error bounds for the approximation of distributions. The approximation error is quantified by the maximal density ratio of the distribution $Q$ to be approximated and its proxy $P$. This non-symmetric measure is more informative than and implies bounds for the total variation distance. Explicit approximation problems include, among others, hypergeometric by binomial distributions, binomial by Poisson distributions, and beta by gamma distributions. In many cases we provide both upper and (matching) lower bounds.

Recommendations

• scientific article; zbMATH DE number 4184648
• A bound on the Poisson-binomial relative error
• scientific article; zbMATH DE number 503466
• On the bound of proximity of the binomial distribution to the normal one
• A Poisson-binomial relative error uniform bound

Cites work

• A course on point processes
• A dozen de Finetti-style results in search of a theory
• A general Poisson approximation theorem
• A Remark on Stirling's Formula
• Asymptotical independence of the lengths of subintervals of a randomly partitioned interval; a sample from S. Ikeda's work
• Binomial approximation to the Poisson binomial distribution
• Exact bounds on the closeness between the Student and standard normal distributions
• Finite exchangeable sequences
• scientific article; zbMATH DE number 646828 (Why is no real title available?)
• NIST handbook of mathematical functions
• On the Poisson--binomial relative error
• On the rate of convergence of normal extremes
• On the rate of Poisson convergence
• The density ratio of Poisson binomial versus Poisson distributions
• The Poisson Approximation to the Poisson Binomial Distribution
• The rate of convergence in law of the maximum of an exponential sample

Cited in

(5)

• Relative Density-Ratio Estimation for Robust Distribution Comparison
• A Unifying Tutorial on Approximate Message Passing
• Some new inequalities for beta distributions
• Distributed computation for marginal likelihood based model choice
• A new computational framework for log-concave density estimation

Describes a project that uses

Uses Software

• DLMF

This page was built for publication: Bounding distributional errors via density ratios

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2040039)