Hierin wordt het allemaal uitgelegd.
A Mixture Gatekeeping Procedure Based on the
Hommel Test for Clinical Trial Applications
http://www.multxpert.com/doc/bxdt2011.pdf

3 soorten gatekeeping procedures

- parallel
- serial
- tree-structured

tree-structured is een uitbreiding op 'parallel' en 'serial':

Tree gatekeeping procedures
Efficient way to account for hierarchically ordered multiple objectives in clinical trials.
Extend serial and parallel gatekeeping methods

http://www.bassconference.org/PDFs/BASS 2006 Dmitrienko et a...

It is becoming increasingly common to consider designs with multiple endpoints, analyses and objectives in registration studies because additional information on the efficacy and safety profiles of an experimental drug helps patients, prescribing physicians and payers better understand its properties. More complicated study designs give rise to more sophisticated analysis methods. In the area of multiple comparisons, these problems motivated research on novel testing strategies for hierarchically ordered objectives.

This paper gives an overview of recent developments in this area with emphasis on testing strategies for multiple families of analyses.
The analyses are often related to multiple endpoints (both primary and secondary) but can also represent dose–control comparisons, noninferiority and superiority tests or inferences at several time points. Testing strategies considered here are commonly referred to as gatekeeping strategies. This terminology highlights the fact that the families of analyses are examined sequentially and each one serves as a gatekeeper for the subsequent families. The sequential testing approach reflects the hierarchical nature of the problem and improves the power of the more important analyses placed early in the sequence.

The paper is organized as follows. Section 2 introduces the basic concepts about gatekeeping and establishes notation. Section 3 describes serial gatekeeping procedures and Section 4 describes parallel gatekeeping procedures. A generalization of the two approaches, termed the tree gatekeeping approach, is considered in Section 5.
Tree gatekeeping procedures enable clinical trial researchers to construct testing strategies that combine serial and parallel gatekeepers and also account for logical restrictions among multiple analyses conducted in a clinical trial.

http://users.iems.northwestern.edu/~ajit/papers/40) Gatekeep...

'gatekeeping procedures' zijn gerelateerd aan het 'multiple comparions problem' (zie boven; in het Nederlands 'kanskapitalisatie') --> :

In statistics, the multiple comparisons, multiplicity or multiple testing problem occurs when one considers a set of statistical inferences simultaneously or infers a subset of parameters selected based on the observed values

http://en.wikipedia.org/wiki/Multiple_comparisons_problem

In Dutch: kanskapitalisatie

http://nl.wikipedia.org/wiki/Kanskapitalisatie

Er zijn vele correctiemethoden voor kanskapitalisatie. Een bekende methode is de Bonferroni-correctie.

http://nl.wikipedia.org/wiki/Kanskapitalisatie

De Bonferroni-correctie (of Bonferroni-procedure) is een statistische methode ter bestrijding van het probleem van kanskapitalisatie.

http://nl.wikipedia.org/wiki/Bonferroni-correctie


Wanneer moet ik corrigeren voor multiple testing?

http://os1.amc.nl/wikistatistiek/index.php?title=Multiple_te...


gatekeeping procedure

http://www.proz.com/kudoz/english_to_dutch/medical_general/4...