tgp: Bayesian treed Gaussian process models

Bayesian nonstationary, semiparametric nonlinear regression and design by treed Gaussian processes (GPs) with jumps to the limiting linear model (LLM). Special cases also implemented include Bayesian linear models, CART, treed linear models, stationary separable and isotropic GPs, and GP single-index models. Provides 1-d and 2-d plotting functions (with projection and slice capabilities) and tree drawing, designed for visualization of tgp-class output. Sensitivity analysis and multi-resolution models are supported. Sequential experimental design and adaptive sampling functions are also provided, including ALM, ALC, and expected improvement. The latter supports derivative-free optimization of noisy black-box functions.

Version: 2.4-9
Depends: R (≥ 2.14.0)
Suggests: akima, maptree, MASS
Published: 2013-04-04
Author: Robert B. Gramacy and Matt A. Taddy
Maintainer: Robert B. Gramacy <rbgramacy at>
License: LGPL-2 | LGPL-2.1 | LGPL-3 [expanded from: LGPL]
NeedsCompilation: yes
Citation: tgp citation info
Materials: README ChangeLog
In views: Bayesian, ExperimentalDesign, MachineLearning, Spatial
CRAN checks: tgp results


Reference manual: tgp.pdf
Vignettes: a guide to the tgp package
new features in tgp version 2.x
Package source: tgp_2.4-9.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: tgp_2.4-9.tgz, r-oldrel: tgp_2.4-9.tgz
OS X Mavericks binaries: r-release: tgp_2.4-9.tgz
Old sources: tgp archive

Reverse dependencies:

Reverse depends: earlywarnings, penalizedSVM, plgp, RNCEP
Reverse imports: bootfs, c060, laGP
Reverse suggests: dynaTree, SPOT