The effect of assigning different index dates for control exposure measurement on odds ratio estimates
In case-control studies it is reasonable to consider the exposure history of a case prior to disease onset. For the controls, it is necessary to define comparable periods of exposure opportunity. Motivated by data from a case-control study of the environmental risk factors for Multiple Sclerosis we...
Main Author: | Lundy, Erin |
---|---|
Other Authors: | David B Wolfson (Internal/Supervisor) |
Format: | Others |
Language: | en |
Published: |
McGill University
2012
|
Subjects: | |
Online Access: | http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=110733 |
Similar Items
-
A study of non-collapsibility of the odds ratio via marginal structural and logistic regression models
by: Pang, Menglan
Published: (2012) -
Targeted maximum likelihood estimation for longitudinal data
by: Schnitzer, Mireille
Published: (2013) -
Separable variance-covariance structure: estimation, testing and environmental application
by: Manceur, Marc Ameur
Published: (2012) -
Common pitfalls in statistical analysis: Odds versus risk
by: Priya Ranganathan, et al.
Published: (2015-01-01) -
Functional data classification and covariance estimation
Published: (2011)