The Completion of Covariance Kernels

TL;DR

This paper develops a comprehensive theory for extending partial covariance kernels on serrated domains to full domains, including explicit constructions, conditions for uniqueness, and methods for estimation from data.

Contribution

It introduces a complete framework for positive-semidefinite continuation on serrated domains, including explicit canonical completions and conditions for uniqueness.

Findings

Canonical completion always exists and can be explicitly constructed.

All possible completions are perturbations of the canonical completion.

Estimation reduces to solving linear inverse problems with convergence rates.

Abstract

We consider the problem of positive-semidefinite continuation: extending a partially specified covariance kernel from a subdomain $\Omega$ of a rectangular domain $I\times I$ to a covariance kernel on the entire domain $I\times I$. For a broad class of domains $\Omega$ called \emph{serrated domains}, we are able to present a complete theory. Namely, we demonstrate that a canonical completion always exists and can be explicitly constructed. We characterise all possible completions as suitable perturbations of the canonical completion, and determine necessary and sufficient conditions for a unique completion to exist. We interpret the canonical completion via the graphical model structure it induces on the associated Gaussian process. Furthermore, we show how the estimation of the canonical completion reduces to the solution of a system of linear statistical inverse problems in the space of Hilbert-Schmidt operators, and derive rates of convergence. We conclude by providing extensions of our theory to more general forms of domains, and by demonstrating how our results can be used to construct covariance estimators from sample path fragments of the associated stochastic process. Our results are illustrated numerically by way of a simulation study and a real example.