Orthogonal Gaussian process models

DOI10.5705/SS.202015.0404zbMath1390.62047arXiv1611.00203OpenAlexW2963525373MaRDI QIDQ4639565

Publication date: 9 May 2018

Published in: Unnamed Author (Search for Journal in Brave)

Abstract: Gaussian processes models are widely adopted for nonparameteric/semi-parametric modeling. Identifiability issues occur when the mean model contains polynomials with unknown coefficients. Though resulting prediction is unaffected, this leads to poor estimation of the coefficients in the mean model, and thus the estimated mean model loses interpretability. This paper introduces a new Gaussian process model whose stochastic part is orthogonal to the mean part to address this issue. This paper also discusses applications to multi-fidelity simulations using data examples.

Full work available at URL: https://arxiv.org/abs/1611.00203

zbMATH Keywords

kriging computer experiments identifiability universal kriging multifidelity simulations nonparameteric/semiparametric modeling

Mathematics Subject Classification ID

Gaussian processes (60G15) Design of statistical experiments (62K99) Nonparametric estimation (62G05)

Related Items (12)

Bayesian integrative analysis for multi-fidelity computer experiments ⋮ A Review of Spatial Causal Inference Methods for Environmental and Epidemiological Applications ⋮ A framework for machine learning of model error in dynamical systems ⋮ Indicator-based Bayesian variable selection for Gaussian process models in computer experiments ⋮ Modified Penalized Blind Kriging for efficiently selecting a global trend model ⋮ Transformation and Additivity in Gaussian Processes ⋮ Optimal model average prediction in orthogonal kriging models ⋮ A review on computer model calibration ⋮ A generalized Gaussian process model for computer experiments with binary time series ⋮ Adaptive Bayesian nonparametric regression using a kernel mixture of polynomials with application to partial linear models ⋮ Unnamed Item ⋮ Adjustments to Computer Models via Projected Kernel Calibration

This page was built for publication: Orthogonal Gaussian process models