RStan Package.pdf

The `rstan` package is the R interface to the Stan C++ package, providing user-facing R functions to parse, compile, test, estimate, and analyze Stan models. It implements full Bayesian statistical inference via Markov Chain Monte Carlo (MCMC), rough Bayesian inference via variational approximation, and penalized maximum likelihood estimation via optimization.

Key Features

Full Bayesian inference using No-U-Turn sampler (NUTS), a variant of Hamiltonian Monte Carlo (HMC)
Approximate Bayesian inference using automatic differentiation variational inference (ADVI)
Penalized maximum likelihood estimation using L-BFGS optimization
Automatic differentiation for quick and accurate evaluation of gradients

Related Packages

`bayesplot`: ggplot-based plotting of parameter estimates, diagnostics, and posterior predictive checks
`shinystan`: Interactive GUI for exploring MCMC output
`loo`: Out-of-sample predictive performance estimates and model comparison
`rstanarm`: R formula interface for applied regression modeling
`rstantools`: Tools for developers of R packages interfacing with Stan

Authors and Maintainers

Jonah Gabry (author)
Ben Goodrich (maintainer, author)
Jiqiang Guo (author)

Citation

Please use 'Stan Development Team' when citing the R interface.