|
Abstract:
|
In multiple testing problems it is often of interest to test a sequence of hypotheses allowing all the hypotheses a chance to be tested. The existing step-wise FDR controlling procedures, such as the Benjamini-Hochberg procedure, are not applicable here because the testing process stops after a certain threshold is reached. In this research, powerful procedures are developed under various dependencies for controlling FDR that allow testing of all hypotheses in a sequence. Simulation studies demonstrate the effectiveness of these procedures in terms of FDR control and average power.
|
