Robust inference/testing of treatment effects under covariate-adaptive randomization. The package implements common covariate-adaptive randomization schemes, three estimators, that are model-free and generally applicable under many covariate-adaptive randomization methods, including the minimization, stratified permuted block, stratified biased coin, and stratified urn design. The package can be used for estimating/inference on mean potential outcomes and any contrasts.
To install the most up-to-date version, run the following command in R
devtools::install_github("mbannick/RobinCar")
A vignette (still under development) is at https://marlenabannick.com/RobinCar/index.html
References:
- Ting Ye, Yanyao Yi, Jun Shao (2021). Inference on Average Treatment Effect under Minimization and Other Covariate-Adaptive Randomization Methods. Biometrika. https://doi.org/10.1093/biomet/asab015.
- Ting Ye, Jun Shao, Yanyao Yi, Qingyuan Zhao (2021). Toward better practice of covariate adjustment in analyzing clinical trials. JASA https://arxiv.org/abs/2009.11828.