On February 10, 2025, it was announced that Dr. Rianne de Heide is one of the NWO Open Competition Domain Science-M2 programme recipients.
Rianne de Heide has been awarded an NWO grant for her research focusing on 'Flexible and user-adaptive statistical inference’.
In this project we develop a mathematical framework for multiple testing enabling flexibility in study design, analysis, and model choice, while retaining strong error guarantees. This means that researchers are able to adapt their research questions and sampling plans to the data as they come in, and in a way that is as model-free as possible. We achieve this by combining three recent exciting novel developments in mathematical statistics. The new theory of hypothesis testing with e-values, co-developed by me, allows for flexible sampling plans. Jelle Goeman co-developed the closed testing framework that gives flexibility for defining research questions in a data-dependent way. Thirdly, we add permutation- and knockoff-based methods that allow for relatively model-free inference. In this proposal we aim to combine the strengths of these three frameworks into a novel and coherent whole.
