bridgesampling is an R package for conducting Bayesian model comparisons using bridge sampling (Meng & Wong, 1996).
Specifically, it allows one to compute marginal likelihoods, Bayes factors, and posterior model probabilities.
Meng, X.-L., & Wong, W. H. (1996). Simulating ratios of normalizing constants via a simple identity: A theoretical exploration. Statistica Sinica, 6, 831-860.
For additional information, see the vignettes with a hierarchical normal example implemented in Jags (link) and in Stan (link), and a Bayesian t-test example implemented in Stan (link).
-
bridgesamplingis available from CRAN so the current stable version can be installed directly via:install.packages("bridgesampling") -
To install the latest development version you will need the
devtoolspackage:devtools::install_github("quentingronau/bridgesampling@master") -
For building, use
--compact-vignettes="gs+qpdf"
Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.