Salvato in:
Dettagli Bibliografici
Autori principali: Gossard, Florian, Bachoc, François, Baccou, Jean, Gouic, Thibaut Le, Liandrat, Jacques, Glantz, Tony
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2510.21277
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917039523758080
author Gossard, Florian
Bachoc, François
Baccou, Jean
Gouic, Thibaut Le
Liandrat, Jacques
Glantz, Tony
author_facet Gossard, Florian
Bachoc, François
Baccou, Jean
Gouic, Thibaut Le
Liandrat, Jacques
Glantz, Tony
contents This work addresses the interpolation of probability measures within a spatial statistics framework. We develop a Kriging approach in the Wasserstein space, leveraging the quantile function representation of the one-dimensional Wasserstein distance. To mitigate the inaccuracies in semivariogram estimation that arise from sparse datasets, we combine this formulation with cross-validation techniques. In particular, we introduce a variant of the virtual cross-validation formulas tailored to quantile functions. The effectiveness of the proposed method is demonstrated on a controlled toy problem as well as on a real-world application from nuclear safety.
format Preprint
id arxiv_https___arxiv_org_abs_2510_21277
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Kriging measure-valued data with sparse observations: application to nuclear safety studies
Gossard, Florian
Bachoc, François
Baccou, Jean
Gouic, Thibaut Le
Liandrat, Jacques
Glantz, Tony
Statistics Theory
This work addresses the interpolation of probability measures within a spatial statistics framework. We develop a Kriging approach in the Wasserstein space, leveraging the quantile function representation of the one-dimensional Wasserstein distance. To mitigate the inaccuracies in semivariogram estimation that arise from sparse datasets, we combine this formulation with cross-validation techniques. In particular, we introduce a variant of the virtual cross-validation formulas tailored to quantile functions. The effectiveness of the proposed method is demonstrated on a controlled toy problem as well as on a real-world application from nuclear safety.
title Kriging measure-valued data with sparse observations: application to nuclear safety studies
topic Statistics Theory
url https://arxiv.org/abs/2510.21277