Program
12.00 – 12.15 Introduction of the new PhD student
12.15 - 13.00 CELS PhD students Ivar Grytten and Knut Dagestand Rand present their progress
13.00 – 13.15 Lunch break with sandwiches and coffee
13.15 – 14.00 Invited lecture by Dr. Florian Frommelt, entitled "FDR controlling model selection with applications in GWAS"
14.20 – 14.40 CELS and future plans – discussion around future strategy: Kjetill and all
Registration: https://nettskjema.no/a/75093.html
Abstract for the talk by Florian:
Penalized selection criteria like AIC or BIC are among the most popular methods for variable selection. Their theoretical properties have been studied intensively and are well understood in case of a moderate number of variables. However, these criteria do not work well in a high-dimensional setting under the assumption of sparsity. We will introduce different modifications of AIC and BIC which will allow to control the family wise error rate (FWER) or false discovery rate (FDR), respectively, in terms of including false positive regressors in the model. We will describe certain optimality properties of these criteria and specifically discuss the advantages of controlling FDR rather than FWER. Finally we will illustrate the performance of model selection based on FDR controlling criteria in the context of GWAS.