Detecting the direction of a signal on high-dimensional spheres: non-null and Le Cam optimality results

DOI10.1007/S00440-019-00937-XzbMATH Open1436.62195arXiv1711.02504OpenAlexW2965335412MaRDI QIDQ2174669FDOQ2174669

Authors: Davy Paindaveine, Thomas Verdebout 

Publication date: 21 April 2020

Published in: Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete (Search for Journal in Brave)

Abstract: We consider one of the most important problems in directional statistics, namely the problem of testing the null hypothesis that the spike direction $heta$ of a Fisher-von Mises-Langevin distribution on the $p$-dimensional unit hypersphere is equal to a given direction $het{a}_0$. After a reduction through invariance arguments, we derive local asymptotic normality (LAN) results in a general high-dimensional framework where the dimension $p_n$ goes to infinity at an arbitrary rate with the sample size $n$, and where the concentration $kapp{a}_n$ behaves in a completely free way with $n$, which offers a spectrum of problems ranging from arbitrarily easy to arbitrarily challenging ones. We identify various asymptotic regimes, depending on the convergence/divergence properties of $(kapp{a}_n)$, that yield different contiguity rates and different limiting experiments. In each regime, we derive Le Cam optimal tests under specified $kapp{a}_n$ and we compute, from the Le Cam third lemma, asymptotic powers of the classical Watson test under contiguous alternatives. We further establish LAN results with respect to both spike direction and concentration, which allows us to discuss optimality also under unspecified $kapp{a}_n$. To investigate the non-null behavior of the Watson test outside the parametric framework above, we derive its local asymptotic powers through martingale CLTs in the broader, semiparametric, model of rotationally symmetric distributions. A Monte Carlo study shows that the finite-sample behaviors of the various tests remarkably agree with our asymptotic results.

Full work available at URL: https://arxiv.org/abs/1711.02504

Recommendations

• High-dimensional tests for spherical location and spiked covariance
• Inference on the mode of weak directional signals: a Le Cam perspective on hypothesis testing near singularities
• Testing uniformity on high-dimensional spheres against monotone rotationally symmetric alternatives
• Local Powers of Optimal One‐sample and Multi‐sample Tests for the Concentration of Fisher‐von Mises‐Langevin Distributions
• Tests for the mean direction of the Langevin distribution with large concentration parameter

zbMATH Keywords

high-dimensional statisticsinvariancelocal asymptotic normalityrotationally symmetric distributionsLe Cam's asymptotic theory of statistical experiments

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03) Directional data; spatial statistics (62H11)

Cites Work

• Spherical regression for concentrated Fisher-von Mises distributions
• Title not available (Why is that?)
• Asymptotic Statistics
• Testing Statistical Hypotheses
• A distribution-free M-estimator of multivariate scatter
• Asymptotics in statistics. Some basic concepts.
• Title not available (Why is that?)
• Group Invariance in Statistical Inference
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Computation of Modified Bessel Functions and Their Ratios
• Statistics of orthogonal axial frames
• Statistical analysis on high-dimensional spheres and shape spaces
• Distributions of Angles in Random Packing on Spheres
• Optimal R-estimation of a spherical location
• High dimensional limit theorems and matrix decompositions on the Stiefel manifold
• Title not available (Why is that?)
• Estimation of the concentration parameter in von Mises-Fisher distributions
• Skew-symmetric distributions and Fisher information: the double sin of the skew-normal
• Robust direction estimation
• Improved likelihood ratio tests on the von Mises-Fisher distribution
• Title not available (Why is that?)
• Testing uniformity on high-dimensional spheres against monotone rotationally symmetric alternatives
• High-dimensional tests for spherical location and spiked covariance
• Spherical regression
• \(M\)-estimation for location and regression parameters in group models: A case study using Stiefel manifolds
• Inference for spherical location under high concentration
• Inference on the mode of weak directional signals: a Le Cam perspective on hypothesis testing near singularities
• Testing for principal component directions under weak identifiability

Cited In (3)

• Inference for spherical location under high concentration
• Testing uniformity on high-dimensional spheres: the non-null behaviour of the Bingham test
• On the power of axial tests of uniformity on spheres

Uses Software

• movMF

This page was built for publication: Detecting the direction of a signal on high-dimensional spheres: non-null and Le Cam optimality results

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2174669)