Quantitative asymptotics of graphical projection pursuit

DOI10.1214/ECP.V14-1457MaRDI QIDQ1038951zbMATH OpenOpenAlexFDO

Authors Elizabeth Meckes

Publication date 20 November 2009

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

Full work available at URL https://arxiv.org/abs/0811.2769, https://eudml.org/doc/232027, http://www.emis.de/journals/EJP-ECP/_ejpecp/ECP/viewarticle714d.html

zbMATH Keywords

projection pursuitStein's methodconcentration inequalitiesLipschitz distance

Mathematics Subject Classification ID

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

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 $s^{d-1}$.

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

Cited in

(7)

• scientific article; zbMATH DE number 4001189 (Why is no real title available?)
• Approximation of projections of random vectors
• Pointwise Weyl law for graphs from quantized interval maps
• Asymptotics of graphical projection pursuit
• Projections of probability distributions: a measure-theoretic Dvoretzky theorem
• Uniform convergence rates for halfspace depth
• On low-dimensional projections of high-dimensional distributions

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)