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).

• bridgesampling is 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 devtools package: 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.