(Non)linear multivariate, multilevel modeling via Stan
The brms package provides a flexible interface to fit Bayesian generalized (non)linear multivariate multilevel models using Stan.
brms allows users to specify models via the customary R commands, where
models are specified with formula syntax,
data is provided as a data frame, and
additional arguments are available to specify priors and additional structure.
Estimation may be carried out with Markov chain Monte Carlo or variational inference using Stan programs generated on the fly and compiled. Graphical posterior predictive checking, leave-one-out cross-validation, and posterior visualization are tightly integrated.
Visit the source repository github/paul-buerkner/brms (GitHub) for vignettes/tutorials, function documentation, and other information about the package.
Download and Get Started
Instructions for downloading, installing, and getting started with brms on all platforms.
brms’s reference manual and vignettes are also available from CRAN.
Source Code and Issue Tracker
brms’s source code and issue tracker are hosted by GitHub.
brms is open-source licensed under the