Measures of goodness of fit obtained by almost-canonical transformations on Riemannian manifolds

DOI10.1016/J.JMVA.2019.104579zbMATH Open1436.62726arXiv1811.04866OpenAlexW2995070253WikidataQ115345763 ScholiaQ115345763MaRDI QIDQ2293544FDOQ2293544

Authors: A. Kume, P. E. Jupp 

Publication date: 5 February 2020

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Abstract: The standard method of transforming a continuous distribution on the line to the uniform distribution on the unit interval is the probability integral transform. Analogous transforms exist on compact Riemannian manifolds, in that, for each distribution with continuous positive density, there is a continuous mapping of the manifold to itself that transforms the distribution into the uniform distribution. In general, this mapping is far from unique. We introduce a construction of a version of such a probability integral that is almost canonical. The construction is extended to shape spaces, Cartan-Hadamard manifolds, and simplices. The probability integral transform is used to derive tests of goodness of fit from tests of uniformity. Illustrative examples of these tests of goodness of fit are given involving (i) Fisher distributions on the 2-sphere, (ii) isotropic Mardia-Dryden distributions on the shape space of triangles in the plane. Their behaviour is investigated by simulation.

Full work available at URL: https://arxiv.org/abs/1811.04866

Recommendations

• Sobolev tests of goodness of fit of distributions on compact Riemannian manifolds
• A test of uniformity on shape spaces.
• Data-driven tests of uniformity on product manifolds
• Data-driven Sobolev tests of uniformity on compact Riemannian manifolds
• Multivariate goodness-of-fit on flat and curved spaces via nearest neighbor distances

zbMATH Keywords

compositional datadirectional statisticssimplexCartan-Hadamard manifoldprobability integral transformexponential mapshape space

Mathematics Subject Classification ID

Statistics on manifolds (62R30) Probabilistic measure theory (60A10)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Analysis of principal nested spheres
• Title not available (Why is that?)
• Remarks on a Multivariate Transformation
• Statistical shape analysis. With applications in R
• Smooth tests of goodness-of-fit for directional and axial data
• Title not available (Why is that?)
• On the Volume Elements on a Manifold
• Riemannian geometry
• Title not available (Why is that?)
• Title not available (Why is that?)
• The statistical theory of shape
• Invariant tests for uniformity on compact Riemannian manifolds based on Sobolev norms
• Parametric bootstrap goodness-of-fit testing for Wehrly-Johnson bivariate circular distributions
• On a class of circulas: copulas for circular distributions
• Exponential models for directional data
• Möbius deconvolution on the hyperbolic plane with application to impedance density estimation
• Copulae on products of compact Riemannian manifolds
• Sobolev tests of goodness of fit of distributions on compact Riemannian manifolds
• Directional statistics and shape analysis
• Title not available (Why is that?)
• Significance points for some tests of uniformity on the sphere^†

Cited In (9)

• On a projection-based class of uniformity tests on the hypersphere
• Recent advances in directional statistics
• A test of uniformity on shape spaces.
• Spatial quantiles on the hypersphere
• Sobolev tests of goodness of fit of distributions on compact Riemannian manifolds
• Tractable circula densities from Fourier series
• Data-driven stabilizations of goodness-of-fit tests
• Rates of the strong uniform consistency for the kernel-type regression function estimators with general kernels on manifolds
• Conditional and predictive probability integral transforms

This page was built for publication: Measures of goodness of fit obtained by almost-canonical transformations on Riemannian manifolds

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2293544)