Modeling large nonstationary spatial data with the full-scale basis graphical lasso
Spatial Statistics · 2026-08 · 74 · 101007
DOI: 10.1016/j.spasta.2026.101007 ↗Kurz gefasst
Fachartikel in der Zeitschrift Spatial Statistics, 2026-08. Verfasst von Matthew LeDuc und 2 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Gefördert von: NASA, National Science Foundation.
Abstract (Originalsprache)
We propose a new approach for the modeling large datasets of nonstationary spatial processes that combines a latent low rank process and a sparse covariance model. The low rank component coefficients are endowed with a flexible graphical Gaussian Markov random field model. The utilization of a low rank and compactly-supported covariance structure combines the full-scale approximation and the basis graphical lasso; we term this new approach the full-scale basis graphical lasso (FSBGL). Estimation employs a graphical lasso-penalized likelihood, which is optimized using a difference-of-convex scheme. We illustrate the proposed approach on synthetic fields as well as with a challenging high-resolution simulation dataset of the thermosphere. In a comparison against state-of-the-art spatial models, the FSBGL performs better at capturing salient features of the thermospheric temperature fields, even with limited available training data.
Förderung
- NASA
- National Science Foundation
Zitieren
APA
LeDuc, M., Kleiber, W., & Matsuo, T. (2026). Modeling large nonstationary spatial data with the full-scale basis graphical lasso. Spatial Statistics, 74, 101007. https://doi.org/10.1016/j.spasta.2026.101007
Vancouver
LeDuc M, Kleiber W, Matsuo T. Modeling large nonstationary spatial data with the full-scale basis graphical lasso. Spatial Statistics. 2026;74:101007. doi:10.1016/j.spasta.2026.101007
BibTeX
@article{LeDuc2026,
author = {LeDuc, Matthew and Kleiber, William and Matsuo, Tomoko},
title = {Modeling large nonstationary spatial data with the full-scale basis graphical lasso},
journal = {Spatial Statistics},
year = {2026},
volume = {74},
pages = {101007},
doi = {10.1016/j.spasta.2026.101007}
}
EUVIMED