Hi all,
I'm analyzing a set of longitudinal data obtained from subjects at regular time intervals following an intervention. I want to look for the effect of time, but I also want to adjust the observed time effect for age and sex of subjects. My initial thought was to run a linear model like this:
lm(Signal ~ SubjectID + Time + Age + Sex)
However, age and sex are both attributes of subjectID, and, therefore, are not independent covariates. Should I rather simplify the model like this:
lm(Signal ~ SubjectID + Time) ?
Any suggestions are much appreciated.
Thank you very much!
one question. What is SubjectID? Are they numerical measurements?
Although I suggest using the model includes age and sex
You need something more than standard linear models to analyse longitudinal data. You dependent variable will be time varying (and probably temporally autocorrelated) but sex will be time invariant and apply to the person and not the occasion . The natural method of analysis is the two level random effects model can handle within-individual and between-individual variation.
This is a short practical example of what can be done:
http://www.bristol.ac.uk/media-library/sites/cmm/migrated/documents/repeated-measures.pdf
this is an overview of alternative models (there is a large debate between fixed and random effects models - the former cannot handle time invariant variables)
Article Fixed and Random effects models: making an informed choice
this is a comprehensive site that considers a range of software :
https://www.pilesofvariance.com/
There is much more to say ... for example you may wish to interact your intervention dummy with Sex to see if the treatments is differential effective for men and women.
See the attached screenshot for the course notes that will explain @Kelvyn post. With examples and and computer code. David Booth shout-out to Prof Marie Davidian
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression