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 three methods (BSL, uBSL and semiBSL) and two shrinkage estimations (graphical lasso and Warton's estimation). uBSL (Price et al. (2018) <doi:10.1080/10618600.2017.1302882>) uses an unbiased estimator to the normal density. A semi-parametric version of BSL (semiBSL, An et al. (2018) <arXiv:1809.05800>) is more robust to non-normal summary statistics. Shrinkage estimations can help to bring down the number of simulations when the dimension of the summary statistic is high (e.g., BSLasso, An et al. (2019) <doi:10.1080/10618600.2018.1537928>). Extensions to this package are planned.

Version: 3.0.0
Depends: R (≥ 3.4.0)
Imports: glasso, ggplot2, MASS, mvtnorm, copula, graphics, gridExtra, foreach, coda, Rcpp, methods
LinkingTo: Rcpp, RcppArmadillo
Suggests: elliplot, doParallel
Published: 2019-07-10
Author: Ziwen An ORCID iD [aut, cre], Leah F. South ORCID iD [aut], Christopher C. Drovandi ORCID iD [aut]
Maintainer: Ziwen An < at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
Materials: NEWS
CRAN checks: BSL results


Reference manual: BSL.pdf
Package source: BSL_3.0.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release: BSL_3.0.0.tgz, r-oldrel: BSL_3.0.0.tgz
Old sources: BSL archive


Please use the canonical form to link to this page.