|
Abstract:
|
The Hochberg procedure and Benjamini-Hochberg procedure are widely applied in confirmatory clinical studies and exploratory research for multiplicity adjustment. A common misconception is that these procedures control the type I error rate properly if the test statistics are independent or positively correlated. In fact, a much stronger positive dependence assumption needs to be satisfied to guarantee the type I error rate control. In this presentation, we will review several families of dependence structures of random variables in multiple testing procedures, and investigate the type I error rate control of the Simes test under different dependence structures. In addition, a couple of tests to validate if a sample of multivariate data come from a population that satisfies the specific positive dependence condition will be discussed with application to the relation between the progression-free survival (PFS) and overall survival (OS) in metastatic breast cancer patients.
|
