Hi friends,
I have two independent variables I and J, a dependent variable D and a subject variable S. Each S has several repeated I-J-D measurements (as separate records). I did repeated measures regression using PROC MIXED (in SAS). Both I and J appeared to have significant effects on D (though with different p-values) and I obtained regression coefficients for both I and J on D; variable S does not at all occur there, it only defined the subjects with repeated measurements. So far, so good.
Now I want to establish some relation strength between J and D; in general that should be a correlation coefficient. Given the influence of I, it should be a partial correlation coefficient partialling out the influence of I. But I also want to control for the influence of the subject factor S (several observations are repeated measurements on same subjects) in the same analysis. I did not find any solution using PROC CORR, while performing partial correlation.
Any ideas?
Regards - jim.
--
. . . . . . . . . . . . . . . .
Jim Groeneveld, MSc.
Biostatistician
Science Team
Vitatron B.V.
Meander 1051
6825 MJ Arnhem
Tel: +31/0 26 376 7365
Fax: +31/0 26 376 7305
[log in to unmask]
www.vitatron.com
[common disclaimer]
|