This repository contains source files required to compile the RMarkdown notebook that accompanies the following talk at StanCon 2019:
- Bayesian analyses of time-to-event data using the rstanarm R package by Eren M. Elçi and Sam Brilleman
A compiled PDF version of the notebook is included in the repository's root directory. To compile the notebook yourself, you will need to run the RMarkdown file in the root directory, and have the following R packages installed (including their dependencies): knitr, tidyverse, cowplot, bayesplot, survival, rstanarm, loo, pec and survRM2.
Note: At the time of writing this, the version of rstanarm that is required is the development version available here since this contains the stan_surv
modelling function, however, this will be uploaded to CRAN in the near future, so it may be worth checking here whether rstanarm version >=2.19.0 has been released on CRAN. If it has, then it would be easier to install the package directly from within R by calling install.packages("rstanarm")
instead of downloading the development version from GitHub.