[R] ordinal logistic regression for longitudinal data set

[David Winsemius]

On Apr 24, 2009, at 5:22 PM, joy_stat wrote:

>
>
> Hi,
>
>     Can one tell me which procedure will fit an ordinal logistic  
> regression
> model for longitudinal data set.
>
>   To be precise, I have both dichotomous and polytomous items. Also, I
> would like to specify different covariance structures (unstructured,  
> ar1
> etc) for trial runs.
>
That may seem precise to you .... Perhaps you will find useful  
material in chapter 11 of Agresti's Categorical Data Analysis:
http://www.amazon.com/Categorical-Analysis-Wiley-Probability-Statistics/dp/0471360937
http://www.stat.ufl.edu/~aa/cda/cda.html

... and the R/S accompaniment by Thomson:
https://home.comcast.net/~lthompson221/Splusdiscrete2.pdf

She offers worked examples using functions from package geepack.

-- 

David Winsemius, MD
Heritage Laboratories
West Hartford, CT