Inconsistency between conditional and marginal analyses

Conditional and marginal analyses are widely used in clinical studies. However, the results of these two methods may occasionally contradict each other. For instance, marginal analysis may show that the treatment group outperforms the control group, while conditional analysis may suggest the opposit...

Full description

Saved in:
Bibliographic Details
Published in:Biometrical letters 2024-12, Vol.61 (2), p.137-146
Main Authors: Feng, Chang, Liu, Honghong, Wang, Hongyue, Feng, Changyong
Format: Article
Language:English
Subjects:
Conditional analysis
Conditional expectation
Marginal analysis
Simpson’s paradox
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Conditional and marginal analyses are widely used in clinical studies. However, the results of these two methods may occasionally contradict each other. For instance, marginal analysis may show that the treatment group outperforms the control group, while conditional analysis may suggest the opposite. We examine the causes of this inconsistency and provide general sufficient conditions for ensuring consistency.
ISSN:2199-577X
1896-3811
2199-577X
DOI:10.2478/bile-2024-0009