Saved in:
Bibliographic Details
Main Authors: Deng, Linsui, Wu, C. F. Jeff
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.16328
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911454217633792
author Deng, Linsui
Wu, C. F. Jeff
author_facet Deng, Linsui
Wu, C. F. Jeff
contents Computer experiments involving both qualitative and quantitative (QQ) factors have attracted increasing attention. Gaussian process (GP) models have proven effective in this context by choosing specialized covariance functions for QQ factors. In this work, we extend the latent variable-based GP approach, which maps qualitative factors into a continuous latent space, by establishing a general framework to apply standard kernel functions to continuous latent variables. This approach provides a novel perspective for interpreting some existing GP models for QQ factors and introduces new covariance structures in some situations. The ordinal structure can be incorporated naturally and seamlessly in this framework. Furthermore, the Bayesian information criterion and leave-one-out cross-validation are employed for model selection and model averaging. The performance of the proposed method is comprehensively studied on several examples.
format Preprint
id arxiv_https___arxiv_org_abs_2602_16328
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A general framework for modeling Gaussian process with qualitative and quantitative factors
Deng, Linsui
Wu, C. F. Jeff
Methodology
Computer experiments involving both qualitative and quantitative (QQ) factors have attracted increasing attention. Gaussian process (GP) models have proven effective in this context by choosing specialized covariance functions for QQ factors. In this work, we extend the latent variable-based GP approach, which maps qualitative factors into a continuous latent space, by establishing a general framework to apply standard kernel functions to continuous latent variables. This approach provides a novel perspective for interpreting some existing GP models for QQ factors and introduces new covariance structures in some situations. The ordinal structure can be incorporated naturally and seamlessly in this framework. Furthermore, the Bayesian information criterion and leave-one-out cross-validation are employed for model selection and model averaging. The performance of the proposed method is comprehensively studied on several examples.
title A general framework for modeling Gaussian process with qualitative and quantitative factors
topic Methodology
url https://arxiv.org/abs/2602.16328