hierbase: Enabling Hierarchical Multiple Testing
Implementation of hierarchical inference based on Meinshausen (2008).
Hierarchical testing of variable importance. Biometrika, 95(2), 265-278 and
Renaux, Buzdugan, Kalisch, and Bühlmann, (2020). Hierarchical inference for
genome-wide association studies: a view on methodology with software.
Computational Statistics, 35(1), 1-40.
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 a hierarchical multiple testing
correction and controls the family-wise error rate (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.
Please use the canonical form
to link to this page.