Permutation p-value approximation

In this PhD project, I develop an approach for accurate p-value approximation in permutation testing using the GPD (Generalized Pareto Distribution) while strictly avoiding zero p-values. The proposed p-value approximation framework, including precise fitting of GPD parameters, tail threshold detection, and multiple testing adjustment, will be made available in the open-source R package permAprox on GitHub and CRAN.'

The project is carried out under the supervision of Anne-Laure Boulesteix and Christian L. Müller, as well as in collaboration with the group of Prof.Dr. Erika von Mutius at the Helmholtz Zentrum München.