TEASE: Targeted elastic spatial envelope
Abstract
Envelope regression is a crucial part of multivariate linear modeling, leveraging separation into material and immaterial parts to provide parsimonious data reduction. Rekabdarkolaee et al. (2020) construct spatial envelope, a novel multivariate Gaussian process, extended by May et al. (2022) to linear coregionalization envelope. Our proposal (TEASE) combines spatial envelope and full-scale basis graphical lasso (LeDuc et al., 2025), leveraging targeted graphical elastic net (Kovács et al., 2021) for immaterial precision matrices. Material part is a multivariate multiresolution Gaussian Markov random field (Kleiber et al., 2019; Caringi and Secchi, 2026). Enhanced envelope penalty (Kwon and Zou, 2025) regularizes regression coefficients. Demonstrations are on climate data.
Related
- An abelian envelope without the quotient property
- The lower envelope of ultraspherical polynomials
- Computing the convex envelope of bivariate piecewise linear-quadratic functions in linear time
- The injective envelope of simple modules over Leavitt path algebras I: simple left ideals
- A Gamma envelope and sharp moment inequalities for Gaussian quadratic forms
BibTeX
@misc{krock2026tease,
title = {{TEASE: Targeted elastic spatial envelope}},
author = {Mitchell L. Krock},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.08157},
eprint = {2610.08157},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.08157},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - TEASE: Targeted elastic spatial envelope
AU - Mitchell L. Krock
PY - 2026
DA - 2026/10/06
PB - arXiv
DO - 10.48550/arxiv.2610.08157
UR - https://arxiv.org/abs/2610.08157
AB - Envelope regression is a crucial part of multivariate linear modeling, leveraging separation into material and immaterial parts to provide parsimonious data reduction. Rekabdarkolaee et al. (2020) construct spatial envelope, a novel multivariate Gaussian process, extended by May et al. (2022) to linear coregionalization envelope. Our proposal (TEASE) combines spatial envelope and full-scale basis graphical lasso (LeDuc et al., 2025), leveraging targeted graphical elastic net (Kovács et al., 2021) for immaterial precision matrices. Material part is a multivariate multiresolution Gaussian Markov random field (Kleiber et al., 2019; Caringi and Secchi, 2026). Enhanced envelope penalty (Kwon and Zou, 2025) regularizes regression coefficients. Demonstrations are on climate data.
ER -