March 17th 2015
Faculty of Economic and Business Sciences | Salón de Grados
Some Step-up procedures that control the false discovery rate for discrete tests
2015/03/17 – 13:00 h | Sebastian Doehler, Darmstadt University of Applied Sciences
Abstract
In many multiple testing problems control of the false discovery rate is desired. In this talk we develop some step-up procedures that control the false discovery rate under arbitrary dependence of the p-values. When the test statistics are discrete, these new procedures can provide improvements over classical procedures like the Benjamini-Yekutieli procedure by incorporating discreteness in a straightforward way. We illustrate these new procedures with an application to pharmacovigilance data.
Gallery
