Quantitative asymptotics of graphical projection pursuit

DOI10.1214/ECP.V14-1457zbMATH Open1189.60046arXiv0811.2769OpenAlexW2963750764MaRDI QIDQ1038951FDOQ1038951

Authors: Elizabeth Meckes 

Publication date: 20 November 2009

Published in: Electronic Communications in Probability (Search for Journal in Brave)

Abstract: There is a result of Diaconis and Freedman which says that, in a limiting sense, for large collections of high-dimensional data most one-dimensional projections of the data are approximately Gaussian. This paper gives quantitative versions of that result. For a set of deterministic vectors ${x_i}_{i=1}^n$ in $R^d$ with $n$ and $d$ fixed, let $hetain{s}^{d-1}$ be a random point of the sphere and let $m{u}_n^{h}eta$ denote the random measure which puts mass $frac1n$ at each of the points $inprod{x}_{1}heta,\mathrm{...},inprod{x}_{n}heta$. For a fixed bounded Lipschitz test function $f$, $Z$ a standard Gaussian random variable and $sigm{a}^2$ a suitable constant, an explicit bound is derived for the quantity $dsP[|intfdm{u}_n^{h}eta-Ef(sigmaZ)|>epsilon]$. A bound is also given for $dsP[d_{BL}(m{u}_n^{h}eta,N(0,sigm{a}^2))>epsilon]$, where $d_{BL}$ denotes the bounded-Lipschitz distance, which yields a lower bound on the waiting time to finding a non-Gaussian projection of the $x_i$ if directions are tried independently and uniformly on $s^{d-1}$.

Full work available at URL: https://arxiv.org/abs/0811.2769

Recommendations

• Asymptotics of graphical projection pursuit
• Asymptotic optimality of projection methods
• scientific article; zbMATH DE number 4001189
• scientific article; zbMATH DE number 4018127
• Probabilistic pursuits on graphs
• scientific article; zbMATH DE number 66823
• Probabilistic refinement of the asymptotic spectrum of graphs
• Asymptotes and Projections of Convex Sets.
• scientific article; zbMATH DE number 3974053
• Statistical convergence behavior of affine projection algorithms

zbMATH Keywords

projection pursuitStein's methodconcentration inequalitiesLipschitz distance

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Inequalities; stochastic orderings (60E15)

Cited In (7)

• On low-dimensional projections of high-dimensional distributions
• Approximation of projections of random vectors
• Asymptotics of graphical projection pursuit
• Title not available (Why is that?)
• Uniform convergence rates for halfspace depth
• Pointwise Weyl law for graphs from quantized interval maps
• Projections of probability distributions: a measure-theoretic Dvoretzky theorem

This page was built for publication: Quantitative asymptotics of graphical projection pursuit

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1038951)