When is multivariate kriging worthwhile? Design geometry, parameter sharing and response coupling under heterotopic sampling
Abstract
We investigate when multivariate kriging improves on separate prediction, using separable multi-output Gaussian processes as the principal theoretical benchmark. We introduce directed coverage, directed proximity and borrowing potential indices to describe design complementarity beyond exact overlap without response values. Under common radial kernels, we derive finite-design bounds linking geometry to prediction gains with known covariance parameters and information for estimating cross-output dependence. An exact risk decomposition under correct specification separates oracle gains from differences in estimation penalties; we give a mean-square expansion under explicit conditions. We distinguish parameter sharing from response coupling: auxiliary data can improve prediction through shared-scale estimation even when direct borrowing is negligible. This distinction guides a calibrated screening-and-validation procedure comparing separate marginal models, a shared-scale independent model and a coupled model, retaining the shared-scale option when coupling is excluded. In simulations with sparse target data and auxiliary observations retained during validation, the procedure reduces mean latent squared-error risk by 6.66% relative to separate fitting. After calibration, screening reduces runtime by 14.01% relative to direct validation of the same candidates, with similar mean risk. Controlled experiments show that covariance misspecification can reverse fitted gains, while queueing and monitoring applications demonstrate the importance of auxiliary coverage and the prediction task. These findings support task-matched validation and separate assessments of predictive accuracy, uncertainty calibration and computational cost.
Related
- Wasserstein Kernel Regression with Dependent Distribution-Valued Predictors: Local Recurrence, Joint Inference, and Multivariate Gaussian Geometry
- MIMO and Multi-Group Comparison Problem in Process Control: A Multivariate Statistical Framework for Systems Represented with Frequency Response Functions
- When Is Local Time Stepping Worthwhile? A Cost Analysis for Nested Adaptive Mesh Refinement
- Multi-fidelity Monte Carlo estimation of floor response spectra under combined seismic and structural parameter uncertainties
- Closed Response Calculus for SLE Weldings and Weil--Petersson Kähler Geometry
BibTeX
@misc{chen2026when,
title = {{When is multivariate kriging worthwhile? Design geometry, parameter sharing and response coupling under heterotopic sampling}},
author = {Zexun Chen and Jun Fan and Kuo Wang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.06832},
eprint = {2607.06832},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.06832},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - When is multivariate kriging worthwhile? Design geometry, parameter sharing and response coupling under heterotopic sampling
AU - Zexun Chen
AU - Jun Fan
AU - Kuo Wang
PY - 2026
DA - 2026/10/05
PB - arXiv
DO - 10.48550/arxiv.2607.06832
UR - https://arxiv.org/abs/2607.06832
AB - We investigate when multivariate kriging improves on separate prediction, using separable multi-output Gaussian processes as the principal theoretical benchmark. We introduce directed coverage, directed proximity and borrowing potential indices to describe design complementarity beyond exact overlap without response values. Under common radial kernels, we derive finite-design bounds linking geometry to prediction gains with known covariance parameters and information for estimating cross-output dependence. An exact risk decomposition under correct specification separates oracle gains from differences in estimation penalties; we give a mean-square expansion under explicit conditions. We distinguish parameter sharing from response coupling: auxiliary data can improve prediction through shared-scale estimation even when direct borrowing is negligible. This distinction guides a calibrated screening-and-validation procedure comparing separate marginal models, a shared-scale independent model and a coupled model, retaining the shared-scale option when coupling is excluded. In simulations with sparse target data and auxiliary observations retained during validation, the procedure reduces mean latent squared-error risk by 6.66% relative to separate fitting. After calibration, screening reduces runtime by 14.01% relative to direct validation of the same candidates, with similar mean risk. Controlled experiments show that covariance misspecification can reverse fitted gains, while queueing and monitoring applications demonstrate the importance of auxiliary coverage and the prediction task. These findings support task-matched validation and separate assessments of predictive accuracy, uncertainty calibration and computational cost.
ER -