Geometry of sample spaces

DOI10.1016/J.DIFGEO.2023.102029MaRDI QIDQ6175775zbMATH OpenFDO

Authors Philipp Harms, Peter W. Michor, Xavier Pennec, Stefan Sommer

Publication date 18 August 2023

Published in Differential Geometry and its Applications (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2010.08039

zbMATH Keywords

\(k\)-meanscentral-limit theoremconsistencyFréchet meansgeometric statisticsstatistics on metric spacesWasserstein geometry

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Statistics on metric spaces (62R20)

Abstract: In statistics, independent, identically distributed random samples do not carry a natural ordering, and their statistics are typically invariant with respect to permutations of their order. Thus, an $n$-sample in a space $M$ can be considered as an element of the quotient space of $M^n$ modulo the permutation group. The present paper takes this definition of sample space and the related concept of orbit types as a starting point for developing a geometric perspective on statistics. We aim at deriving a general mathematical setting for studying the behavior of empirical and population means in spaces ranging from smooth Riemannian manifolds to general stratified spaces. We fully describe the orbifold and path-metric structure of the sample space when $M$ is a manifold or path-metric space, respectively. These results are non-trivial even when $M$ is Euclidean. We show that the infinite sample space exists in a Gromov-Hausdorff type sense and coincides with the Wasserstein space of probability distributions on $M$. We exhibit Fr'echet means and $k$-means as metric projections onto 1-skeleta or $k$-skeleta in Wasserstein space, and we define a new and more general notion of polymeans. This geometric characterization via metric projections applies equally to sample and population means, and we use it to establish asymptotic properties of polymeans such as consistency and asymptotic normality.

Cites work

• A Central Limit Theorem for Sums of Interchangeable Random Variables
• A course in metric geometry
• A smeary central limit theorem for manifolds with application to high-dimensional spheres
• A Unified View of the Theory of Directional Statistics, 1975-1988
• An invitation to Alexandrov geometry. CAT(0) spaces
• An invitation to statistics in Wasserstein space
• Asymptotic Statistics
• Central Limit Theorems for Interchangeable Processes
• Finite exchangeable sequences
• Geometrical smeariness -- a new phenomenon of Fréchet means
• Geometry of the space of phylogenetic trees
• Hamilton-Jacobi equations and distance functions on Riemannian manifolds
• scientific article; zbMATH DE number 3642411 (Why is no real title available?)
• scientific article; zbMATH DE number 1194132 (Why is no real title available?)
• scientific article; zbMATH DE number 3719944 (Why is no real title available?)
• scientific article; zbMATH DE number 3551577 (Why is no real title available?)
• scientific article; zbMATH DE number 1324235 (Why is no real title available?)
• scientific article; zbMATH DE number 1076783 (Why is no real title available?)
• scientific article; zbMATH DE number 1992395 (Why is no real title available?)
• scientific article; zbMATH DE number 2051142 (Why is no real title available?)
• scientific article; zbMATH DE number 2107988 (Why is no real title available?)
• scientific article; zbMATH DE number 3340881 (Why is no real title available?)
• scientific article; zbMATH DE number 3053873 (Why is no real title available?)
• Intrinsic inference on the mean geodesic of planar shapes and tree discrimination by leaf growth
• Intrinsic statistics on Riemannian manifolds: Basic tools for geometric measurements
• Large sample theory of intrinsic and extrinsic sample means on manifolds. I
• Large sample theory of intrinsic and extrinsic sample means on manifolds. II.
• Limit theorems for empirical Fréchet means of independent and non-identically distributed manifold-valued random variables
• Means in complete manifolds: uniqueness and approximation
• Metric structures for Riemannian and non-Riemannian spaces. Transl. from the French by Sean Michael Bates. With appendices by M. Katz, P. Pansu, and S. Semmes. Edited by J. LaFontaine and P. Pansu
• Nonparametic estimation of location and dispersion on Riemannian manifolds
• On the Convexity of Injectivity Domains on Nonfocal Manifolds
• On the meaning of mean shape: manifold stability, locus and the two sample test
• On the rate of convergence in Wasserstein distance of the empirical measure
• Overview of object oriented data analysis
• Riemannian \(L^{p}\) center of mass: existence, uniqueness, and convexity
• Riemannian center of mass and mollifier smoothing
• Riemannian geometric statistics in medical image analysis
• Shape Manifolds, Procrustean Metrics, and Complex Projective Spaces
• Stability of a 4th-order curvature condition arising in optimal transport theory
• Sticky central limit theorems on open books
• Stochastic Equations in Infinite Dimensions
• Symmetric Measures on Cartesian Products
• The central limit theorem for exchangeable random variables without moments
• The convergence problem in mean field games with local coupling
• Topics in differential geometry
• Variational convergence over metric spaces
• Welcome to Riemannian computing in computer vision

Cited in

(2)

• Coarse embeddings of quotients by finite group actions
• Controllability on landmark manifolds for shapes and neural ODEs

This page was built for publication: Geometry of sample spaces

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6175775)