BSL: Bayesian Synthetic Likelihood

Bayesian synthetic likelihood (BSL, Price et al. (2018) <doi:10.1080/10618600.2017.1302882>) is an alternative to standard, non-parametric approximate Bayesian computation (ABC). BSL assumes a multivariate normal distribution for the summary statistic likelihood and it is suitable when the distribution of the model summary statistics is sufficiently regular. This package provides a Metropolis Hastings Markov chain Monte Carlo implementation of BSL, BSLasso and semiBSL. BSL with graphical lasso (BSLasso, An et al. (2018) <https://eprints.qut.edu.au/102263/>) is computationally more efficient when the dimension of the summary statistic is high. A semi-parametric version of BSL (semiBSL, An et al. (2018) <arXiv:1809.05800>) is more robust to non-normal summary statistics. Extensions to this package are planned.

Version: 2.0.0
Depends: R (≥ 3.4.0)
Imports: glasso, ggplot2, MASS, mvtnorm, copula, cvTools, graphics, gridExtra, foreach, coda, Rcpp, methods
LinkingTo: Rcpp, RcppArmadillo
Suggests: elliplot, doParallel
Published: 2019-01-16
Author: Ziwen An ORCID iD [aut, cre], Leah F. South ORCID iD [aut], Christopher C. Drovandi ORCID iD [aut]
Maintainer: Ziwen An <ziwen.an at hdr.qut.edu.au>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
Materials: NEWS
CRAN checks: BSL results

Downloads:

Reference manual: BSL.pdf
Package source: BSL_2.0.0.tar.gz
Windows binaries: r-devel: BSL_2.0.0.zip, r-release: BSL_2.0.0.zip, r-oldrel: BSL_2.0.0.zip
OS X binaries: r-release: BSL_2.0.0.tgz, r-oldrel: BSL_2.0.0.tgz
Old sources: BSL archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=BSL to link to this page.