This is an R package that fits spatial temporal aggregated predictor models using Stan (via the rstan package) for the back-end estimation. The primary target audience is researchers interested in the effect of built environment features (BEFs) on human health, though other applications are possible. See the package’s website for an introduction. Currently count, binomial, and continuous outcomes are supported.


Development Version

To install the current development version from GitHub, first make sure that you can install the rstan package and C++ toolchain by following these instructions.

Once rstan is successfully installed, you can install rstap from GitHub using the devtools package by executing the following in R:

if (!require(devtools)) {

Note that vignettes for this package are separately available from the rstap website.

If installation fails, or you encounter other problems, please let us know by filing an issue.


Both examples and base code are welcome. Whether you’re commiting a case study or a helping me flesh out further functionality of the package. Contact me via atpvyc at umich dot edu if interested.

How to cite this package

Please use the citation associated with the arxiv preprint.


This work was developed with support from NIH grant R01-HL131610 (PI: Sanchez).