MCMChybridGP: Hybrid Markov chain Monte Carlo using Gaussian Processes

Hybrid Markov chain Monte Carlo (MCMC) to simulate from a multimodal target distribution. A Gaussian process approximation makes this possible when derivatives are unknown. The Package serves to minimize the number of function evaluations in Bayesian calibration of computer models using parallel tempering. It allows replacement of the true target distribution in high temperature chains, or complete replacement of the target. Methods used are described in, "Efficient MCMC schemes for Bayesian calibration of computer models", Fielding, Mark, Nott, David J. and Liong Shie-Yui, Technometrics (2010). The authors gratefully acknowledge the support & contributions of the Singapore-Delft Water Alliance (SDWA). The research presented in this work was carried out as part of the SDWA's Multi-Objective Multi-Reservoir Management research programme (R-264-001-272).

Version: 4.3
Depends: MASS
Published: 2011-08-12
Author: Mark J. Fielding
Maintainer: Mark J. Fielding <mfieldin at uow.edu.au>
License: GPL-2
NeedsCompilation: yes
CRAN checks: MCMChybridGP results

Downloads:

Reference manual: MCMChybridGP.pdf
Package source: MCMChybridGP_4.3.tar.gz
OS X binary: MCMChybridGP_4.3.tgz
Windows binary: MCMChybridGP_4.3.zip
Old sources: MCMChybridGP archive