bhetGP: Bayesian Heteroskedastic Gaussian Processes
Performs Bayesian posterior inference for heteroskedastic Gaussian processes.
Models are trained through MCMC including elliptical slice sampling (ESS) of
latent noise processes and Metropolis-Hastings sampling of
kernel hyperparameters. Replicates are handled efficientyly through a
Woodbury formulation of the joint likelihood for the mean and noise process
(Binois, M., Gramacy, R., Ludkovski, M. (2018) <doi:10.1080/10618600.2018.1458625>)
For large data, Vecchia-approximation for faster
computation is leveraged (Sauer, A., Cooper, A., and Gramacy, R.,
(2023), <doi:10.1080/10618600.2022.2129662>). Incorporates 'OpenMP' and
SNOW parallelization and utilizes 'C'/'C++' under the hood.
| Version: |
1.0 |
| Imports: |
grDevices, graphics, stats, doParallel, foreach, parallel, GpGp, GPvecchia, Matrix, Rcpp, mvtnorm, FNN, hetGP, laGP |
| LinkingTo: |
Rcpp, RcppArmadillo |
| Suggests: |
interp |
| Published: |
2025-07-14 |
| Author: |
Parul V. Patil [aut, cre] |
| Maintainer: |
Parul V. Patil <parulvijay at vt.edu> |
| License: |
LGPL-2 | LGPL-2.1 | LGPL-3 [expanded from: LGPL] |
| NeedsCompilation: |
yes |
| Materials: |
README |
| CRAN checks: |
bhetGP results |
Documentation:
Downloads:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=bhetGP
to link to this page.