|
Abstract:
|
Multiple testing problems are often seen in various clinical trials. An appropriate testing procedure should be specified before conducting the trial to deal with the potential inflation of type I error rate due to multiplicity. Numerous approaches have been developed to perform such testing. In this article, we propose a stepwise progressive parametric multiple (SPPM) testing procedure, which considers the joint distribution of the test statistics. The procedure constructs the testing using the products of all the combinations of local individual p-values and the critical values are determined by numerical integrations progressively using the closure principle. We have compared through extensive simulations the performance of the SPPM to several other existing multiple testing procedures, and demonstrate the advantage of using the SPPM procedure, in terms of power, for the certain situations of multiple testing. The method can also take the clinical importance and/or statistical power into consideration when distributing the error rate among the individual hypotheses.
|
