A new class of non-stationary Gaussian fields with general smoothness on metric graphs

TL;DR

This paper introduces a flexible class of non-stationary Gaussian fields on metric graphs, enabling advanced modeling of network data with arbitrary smoothness and efficient Bayesian inference methods.

Contribution

It proposes a new generalized Whittle-Matérn class on metric graphs, extending regularity results and developing a computational approach for large-scale Bayesian inference.

Findings

The new fields accommodate non-stationarity and arbitrary smoothness.

Explicit convergence rates for covariance approximation are derived.

Simulation and traffic data application demonstrate practical effectiveness.

Abstract

The increasing availability of network data has driven the development of advanced statistical models specifically designed for metric graphs, where Gaussian processes play a pivotal role. While models such as Whittle-Mat\'ern fields have been introduced, there remains a lack of practically applicable options that accommodate flexible non-stationary covariance structures or general smoothness. To address this gap, we propose a novel class of generalized Whittle-Mat\'ern fields, which are rigorously defined on general compact metric graphs and permit both non-stationarity and arbitrary smoothness. We establish new regularity results for these fields, which extend even to the standard Whittle-Mat\'ern case. Furthermore, we introduce a method to approximate the covariance operator of these processes by combining the finite element method with a rational approximation of the operator's fractional power, enabling computationally efficient Bayesian inference for large datasets. Theoretical guarantees are provided by deriving explicit convergence rates for the covariance approximation error, and the practical utility of our approach is demonstrated through simulation studies and an application to traffic speed data, highlighting the flexibility and effectiveness of the proposed model class.