Navigation
This page is maintained by the Multiplicity
Advisory Board ( Alex
Dmitrienko). Annotated bibliography project
The annotated bibliography lists publications
(papers, book chapters and sometimes presentations) that deal
with the analysis of multiple outcomes in pharmaceutical applications.
Each publication is accompanied by a brief summary that describes
the proposed statistical methods and their applications to pharmaceutical
research. The purpose of the annotated bibliography is to help
researchers quickly review publications in the area, understand
the main themes/approaches and assess their applicability to a
particular problem encountered in pharmaceutical drug development.
If you would like to contribute to the annotated
bibliography project, we encourage you to post relevant information
in this discussion thread in the BioPharmNet forum.
Multiple tests based on marginal p-values
rely on elementary probability inequalities and thus do not depend
on the joint distribution of test statistics. These multiple testing
procedures are intuitive, easy to apply and widely used in pharmaceutical
applications.
Parametric multiple tests make explicit assumptions
about the joint distribution of the individual test statistics
(for example, the test statistics follow a multivariate normal
or t distribution). These tests are more powerful than multiple
tests based on marginal p-values.
Multi-stage testing procedures play an important
role in clinical trials with multiple families of tests (often
termed gatekeepers) that are carried out in a sequential manner.
Tests for the global null hypothesis of no
treatment effect are used in clinical trials with multiple (co-primary)
outcome variables when clinical trial researchers are interested
in studying the effects of an investigational therapy on all of
the variables. |
