Isotropic covariance functions on graphs and their edges

DOI10.1214/19-AOS1896zbMATH Open1455.62142arXiv1710.01295OpenAlexW3049260668MaRDI QIDQ2215747FDOQ2215747

Authors: Jesper Møller, Ethan B. Anderes, Jakob Gulddahl Rasmussen 

Publication date: 14 December 2020

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: We develop parametric classes of covariance functions on linear networks and their extension to graphs with Euclidean edges, i.e., graphs with edges viewed as line segments or more general sets with a coordinate system allowing us to consider points on the graph which are vertices or points on an edge. Our covariance functions are defined on the vertices and edge points of these graphs and are isotropic in the sense that they depend only on the geodesic distance or on a new metric called the resistance metric (which extends the classical resistance metric developed in electrical network theory on the vertices of a graph to the continuum of edge points). We discuss the advantages of using the resistance metric in comparison with the geodesic metric as well as the restrictions these metrics impose on the investigated covariance functions. In particular, many of the commonly used isotropic covariance functions in the spatial statistics literature (the power exponential, Mat{'e}rn, generalized Cauchy, and Dagum classes) are shown to be valid with respect to the resistance metric for any graph with Euclidean edges, whilst they are only valid with respect to the geodesic metric in more special cases.

Full work available at URL: https://arxiv.org/abs/1710.01295

Recommendations

• Variance and covariance of distributions on graphs
• Isotropic covariance functions on spheres: some properties and modeling considerations
• Gaussian random fields: with and without covariances
• On the use of non-Euclidean distance measures in geostatistics
• Isotropic covariance matrix functions on all spheres

zbMATH Keywords

reproducing kernel Hilbert spacegeodesic metriclinear networkresistance metricparametric classes of covariance functionsrestricted covariance function properties

Mathematics Subject Classification ID

Inference from spatial processes (62M30) Applications of graph theory (05C90) Probabilistic graphical models (62H22) Analysis of variance and covariance (ANOVA) (62J10) Statistics on manifolds (62R30) Hilbert spaces with reproducing kernels (= (proper) functional Hilbert spaces, including de Branges-Rovnyak and other structured spaces) (46E22) Applications of Markov renewal processes (reliability, queueing networks, etc.) (60K20)

Cites Work

• Spatial Analysis along Networks
• The Dagum family of isotropic correlation functions
• Simple tests for the validity of correlation function models on the circle
• Strictly and non-strictly positive definite functions on spheres
• Log Gaussian Cox Processes
• Title not available (Why is that?)
• A moving average approach for spatial statistical models of stream networks
• Title not available (Why is that?)
• Spatial point patterns: methodology and applications with R
• Title not available (Why is that?)
• Correlation theory of stationary and related random functions. Volume II: Supplementary notes and references
• Geometry of cuts and metrics
• Completely monotonic functions
• Title not available (Why is that?)
• l-1 summability of multiple Fourier integrals and positivity
• Title not available (Why is that?)
• Harmonic analysis for resistance forms.
• On alpha-symmetric multivariate distributions
• On \(\alpha\)-symmetric multivariate characteristic functions
• A Hilbert space approach to effective resistance metric

Cited In (16)

• Bifractional Brownian motions on metric spaces
• Regularity and numerical approximation of fractional elliptic differential equations on compact metric graphs
• Nearest-neighbour Markov point processes on graphs with Euclidean edges
• Vector random fields on the probability simplex with metric-dependent covariance matrix functions
• The Matérn model: a journey through statistics, numerical analysis and machine learning
• Necessary and sufficient conditions for asymptotically optimal linear prediction of random fields on compact metric spaces
• Variance and covariance of distributions on graphs
• Variable selection using penalised likelihoods for point patterns on a linear network
• Cox processes driven by transformed Gaussian processes on linear networks -- a review and new contributions
• Point processes on directed linear networks
• Nested covariance functions on graphs with Euclidean edges cross time
• A flexible Clayton-like spatial copula with application to bounded support data
• Space-time covariance models on networks
• Gaussian Whittle-Matérn fields on metric graphs
• Second-order and local characteristics of network intensity functions
• Gaussian field on the symmetric group: prediction and learning

Uses Software

• SANET

This page was built for publication: Isotropic covariance functions on graphs and their edges

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2215747)