R package supporting case crossover models as described in Approximate Bayesian Inference for Case Crossover Models (2020), accepted to Biometrics.
This package is no longer under development. I am focussing on new methods and software, notably the aghq package and methods (github, vignette, methods/theory preprint). Expect a casecrossover2 package which uses this underlying framework to appear in the future!
This R package implements approximate Bayesian inference for case crossover models. The methodology and interface supports linear and non-linear covariate effects and can handle datasets with hundreds of thousands of observations and multiple control days for each subject.
The package is not yet on CRAN. Install using devtools::install_github("awstringer1/casecrossover").
The first version of this package is nearly completed. For details on what can and can't be done currently, check out the
vignette, either by installing the package and typing vignette("casecrossover"), or from the vignettes/ folder in this
repository.