Volume 24, 2020
|Page(s)||252 - 274|
|Published online||23 April 2020|
On the excursion area of perturbed Gaussian fields
Conservatoire National des Arts et Métiers, Paris, EA4629,
292 rue Saint-Martin,
Paris Cedex 03, France.
2 MAP5 UMR CNRS 8145, Université Paris Descartes, 45 rue des Saints-Pères, 75270 Paris Cedex 06, France.
3 Dipartimento di Matematica e Applicazioni, Università degli Studi di Milano-Bicocca Via Roberto Cozzi, 55, 20125 Milano MI, Italy.
* Corresponding author: firstname.lastname@example.org
Accepted: 13 January 2020
We investigate Lipschitz-Killing curvatures for excursion sets of random fields on ℝ2 under a very specific perturbation, namely a small spatial-invariant random perturbation with zero mean. An expansion formula for mean curvatures is derived when the magnitude of the perturbation vanishes, which recovers the Gaussian Kinematic Formula at the limit by contiguity of the model. We develop an asymptotic study of the perturbed excursion area behaviour that leads to a quantitative non-Gaussian limit theorem, in Wasserstein distance, for fixed small perturbations and growing domain. When letting both the perturbation vanish and the domain grow, a standard Central Limit Theorem follows. Taking advantage of these results, we propose an estimator for the perturbation variance which turns out to be asymptotically normal and unbiased, allowing to make inference through sparse information on the field.
Mathematics Subject Classification: 60G60 / 60F05 / 60G15 / 62M40 / 62F12
Key words: LK curvatures / Gaussian fields / perturbed fields / quantitative limit theorems / sojourn times / sparse inference for random fields / spatial-invariant random perturbations
© The authors. Published by EDP Sciences, SMAI 2020
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Initial download of the metrics may take a while.