[Statlist] Reminder: ETH Young Data Science Researcher Seminar Zurich, Virtual Seminar by Kweku Abraham, Université Paris-​Sud

[Maurer Letizia]

Dear all

We are glad to announce the following talk in the virtual ETH Young Data Science Researcher Seminar Zurich

"Optimal false discovery rate control of a nonparametric hidden Markov model multiple testing procedure"  
by Kweku Abraham, Université Paris-​Sud


Time: Friday, 11. December 2020, 15:00-​16:00
Place: Zoom at https://ethz.zoom.us/j/92367940258

Abstract: Multiple testing has become a very important statistical problem in the age of high-​dimensional data sets. Abstractly, the goal is as follows: Given measurements of very many covariates (for example, the nucleotides forming someones DNA), return those covariates which are predictive of some output data (such as health outcomes). A typical design aim for the statistician is to give a procedure with controlled "size", in that the so-​called False Discovery Rate (FDR) is controlled at some target level, while maximising the number of true discoveries. I will explain that a commonly used method based on posterior ("smoothing") probabilities achieves this goal when the covariates have a Markov dependence structure.
Best wishes,

M. Azadkia, Y. Chen, M. Löffler, A. Taeb

Seminar website: https://math.ethz.ch/sfs/news-and-events/young-data-science.html