Estimates pre-compiled regression models using the 'rstan' package, which provides the R interface to the Stan C++ library for Bayesian estimation. Users specify models via the customary R syntax with a formula and data.frame plus some additional arguments for priors.

Sun November 20 2016 (2.13.1)

- Bug fixes

- Fix bug in reloo() if data was not specified
- Fix bug in pp_validate() that was only introduced on GitHub

- New features

- Uses the new bayesplot and rstantools R packages
- The new prior_summary() function can be used to figure out what priors were actually used
- stan_gamm4() is better implemented, can be followed by plot_nonlinear(), posterior_predict (with newdata), etc.
- Hyperparameters (i.e. covariance matrices in general) for lme4 style models are now returned by as.matrix()
- pp_validate() can now be used if optimization or variational Bayesian inference was used to estimate the original model

Fri September 9 2016 (2.12.1)

- Bug fixes

- Fix for bad bug in posterior_predict() when factor labels have spaces in lme4-style models
- Fix when weights are used in Poisson models

- New features

- posterior_linpred() gains an XZ argument to output the design matrix

Fri July 29 2016 (2.11.1)

- Bug fixes

- Requiring manually specifying offsets when model has an offset and newdata is not NULL

- New features

- stan_biglm function that somewhat supports biglm::biglm
- as.array method for stanreg objects

Fri June 24 2016 (2.10.1)

- Bug fixes

- Works with devtools now

- New features

- k_threshold argument to loo() and kfold() for enhanced model comparison
- Ability to use sparse X matrices (slowly) for many models

Tue May 24 2016 (2.9.0-4)

- Serious bug fixes

- posterior_predict with newdata was wrong for ordinal models
- stan_lm() was wrong if there was no intercept
- stan_glmer.fit() would permit models with duplicative group-specific terms but these are usually a mistake on the user's part

- Bug fixes

- posterior_predict with lme4-style models would fail if there were spaces or colons in the levels of the grouping variables
- posterior_predict with ordinal models would output an integer matrix instead of a character matrix

- New features

- pp_validate() function based on the BayesValidate package by Sam Cook
- posterior_vs_prior() function to visualize the effect of conditioning on the data
- Works (again) with R versions back to 3.0.2 (untested though)

Sat Feb 13 2016 (2.9.0-3)

- Serious bug fixes

- Fix problem with models that had group-specific coefficients, which were mislabled. Although the parameters were estimated correctly, users of previous versions of rstanarm should run such models again to obtain correct summaries and posterior predictions. Thanks to someone named Luke for pointing this problem out on stan-users.

- Bug fixes

- Vignettes now view correctly on the CRAN webiste thanks to Yihui Xie
- Fix problem with models without intercepts thanks to Paul-Christian Buerkner
- Fix problem with specifying binomial 'size' for posterior_predict using newdata
- Fix problem with lme4-style formulas that use the same grouping factor multiple times
- Fix conclusion in rstanarm vignette thanks to someone named Michael

- Features

- Group-specific design matrices are kept sparse throughout to reduce memory consumption
- The log_lik() function now has a newdata argument
- New vignette on hierarchical partial pooling

Sat Jan 09 2016 (2.9.0-1) Initial CRAN release