The R-package hierbase offers tools to perform hierarchical inference for one or multiple data sets based on ready-to-use (group) test functions or alternatively a user specified (group) test function. The procedure is based on an efficient hierarchical multiple testing correction and controls the FWER. The functions can easily be run in parallel. Hierarchical inference can be applied to (low- or) high-dimensional data sets to find significant groups or single variables (depending on the signal strength and correlation structure) in a data-driven and automated procedure. Possible applications can for example be found in statistical genetics and statistical genomics.


You can install the development version from Github by running

# install.packages("devtools")


Renaux, C., B├╝hlmann, P. (2021), Efficient Multiple Testing Adjustment for Hierarchical Inference. <arXiv:2104.15028>