English

Convergence of Multivariate Quantile Surfaces

Statistics Theory 2016-12-06 v2 Statistics Theory

Abstract

We define the quantile set of order α[1/2,1)\alpha \in \left[ 1/2,1\right) associated to a law PP on Rd\mathbb{R}^{d} to be the collection of its directional quantiles seen from an observer ORdO\in \mathbb{R}^{d}. Under minimal assumptions these star-shaped sets are closed surfaces, continuous in (O,α)(O,\alpha ) and the collection of empirical quantile surfaces is uniformly consistent.\ Under mild assumptions -- no density or symmetry is required for PP -- our uniform central limit theorem reveals the correlations between quantile points and a non asymptotic Gaussian approximation provides joint confident enlarged quantile surfaces. Our main result is a dimension free rate n1/4(logn)1/2(loglogn)1/4n^{-1/4} (\log n)^{1/2}(\log\log n) ^{1/4} of Bahadur-Kiefer embedding by the empirical process indexed by half-spaces. These limit theorems sharply generalize the univariate quantile convergences and fully characterize the joint behavior of Tukey half-spaces.

Keywords

Cite

@article{arxiv.1607.02604,
  title  = {Convergence of Multivariate Quantile Surfaces},
  author = {Adil Ahidar-Coutrix and Philippe Berthet},
  journal= {arXiv preprint arXiv:1607.02604},
  year   = {2016}
}

Comments

version 2

R2 v1 2026-06-22T14:49:56.293Z