Markov properties of Gaussian random fields on compact metric graphs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bolin, David, Simas, Alexandre B., Wallin, Jonas
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916392972845056
author Bolin, David
Simas, Alexandre B.
Wallin, Jonas
author_facet Bolin, David
Simas, Alexandre B.
Wallin, Jonas
contents There has recently been much interest in Gaussian fields on linear networks and, more generally, on compact metric graphs. One proposed strategy for defining such fields on a metric graph $Γ$ is through a covariance function that is isotropic in a metric on the graph. Another is through a fractional-order differential equation $L^{α/2} (τu) = \mathcal{W}$ on $Γ$, where $L = κ^2 - \nabla(a\nabla)$ for (sufficiently nice) functions $κ, a$, and $\mathcal{W}$ is Gaussian white noise. We study Markov properties of these two types of fields. First, we show that no Gaussian random fields exist on general metric graphs that are both isotropic and Markov. Then, we show that the second type of fields, the generalized Whittle--Matérn fields, are Markov if $α\in\mathbb{N}$, and conversely, if $a$ and $κ$ are constant and $u$ is Markov, then $α\in\mathbb{N}$. Further, if $α\in\mathbb{N}$, a generalized Whittle--Matérn field $u$ is Markov of order $α$, which means that the field $u$ in one region $S\subsetΓ$ is conditionally independent of $u$ in $Γ\setminus S$ given the values of $u$ and its $α-1$ derivatives on $\partial S$. Finally, we provide two results as consequences of the theory developed: first we prove that the Markov property implies an explicit characterization of $u$ on a fixed edge $e$, revealing that the conditional distribution of $u$ on $e$ given the values at the two vertices connected to $e$ is independent of the geometry of $Γ$; second, we show that the solution to $L^{1/2}(τu) = \mathcal{W}$ on $Γ$ can obtained by conditioning independent generalized Whittle--Matérn processes on the edges, with $α=1$ and Neumann boundary conditions, on being continuous at the vertices.
format Preprint
id arxiv_https___arxiv_org_abs_2304_03190
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Markov properties of Gaussian random fields on compact metric graphs
Bolin, David
Simas, Alexandre B.
Wallin, Jonas
Probability
Statistics Theory
60G60 (Primary) 60G15, 60H15 (Secondary)
There has recently been much interest in Gaussian fields on linear networks and, more generally, on compact metric graphs. One proposed strategy for defining such fields on a metric graph $Γ$ is through a covariance function that is isotropic in a metric on the graph. Another is through a fractional-order differential equation $L^{α/2} (τu) = \mathcal{W}$ on $Γ$, where $L = κ^2 - \nabla(a\nabla)$ for (sufficiently nice) functions $κ, a$, and $\mathcal{W}$ is Gaussian white noise. We study Markov properties of these two types of fields. First, we show that no Gaussian random fields exist on general metric graphs that are both isotropic and Markov. Then, we show that the second type of fields, the generalized Whittle--Matérn fields, are Markov if $α\in\mathbb{N}$, and conversely, if $a$ and $κ$ are constant and $u$ is Markov, then $α\in\mathbb{N}$. Further, if $α\in\mathbb{N}$, a generalized Whittle--Matérn field $u$ is Markov of order $α$, which means that the field $u$ in one region $S\subsetΓ$ is conditionally independent of $u$ in $Γ\setminus S$ given the values of $u$ and its $α-1$ derivatives on $\partial S$. Finally, we provide two results as consequences of the theory developed: first we prove that the Markov property implies an explicit characterization of $u$ on a fixed edge $e$, revealing that the conditional distribution of $u$ on $e$ given the values at the two vertices connected to $e$ is independent of the geometry of $Γ$; second, we show that the solution to $L^{1/2}(τu) = \mathcal{W}$ on $Γ$ can obtained by conditioning independent generalized Whittle--Matérn processes on the edges, with $α=1$ and Neumann boundary conditions, on being continuous at the vertices.
title Markov properties of Gaussian random fields on compact metric graphs
topic Probability
Statistics Theory
60G60 (Primary) 60G15, 60H15 (Secondary)
url https://arxiv.org/abs/2304.03190