[R] repeated measures with gls

Celina Zabaloy celina.zabaloy at gmail.com
Mon Feb 2 21:49:50 CET 2009

I am using the gls function of the nlme package to analyze data sets of 
soil respiration which have the following design: 3 complete blocks x 5 
sampling dates (time from fertilization) x 3 fertilization levels. The 
fertilization dates are equal for all subjects (blocks) but not 
periodical (-46, 10, 24, 53, 123 days from the event).
The code that I've been using is:
fit.csnC<- gls(dno.C ~ block+Fertil*factor(Samp), data=dDakot, 
method="ML", corr=corCompSymm(, form=~1|block))

fit.arhnC<-update(fit.csnC, corr=corAR1(, form=~1|block), 
weight=varIdent(form =~1|Samp))
I've plotted the model with :
plot(fit.csnC,dno.C ~ fitted(.), abline = c(0,1))
and residuals look fine, same as observed vs. fitted values.
I would like to make sure that the model specification and the gls 
function are correct for this design. Then, if the model is OK, how do I 
do to test for main and simple effects, ie. how can I apply contrasts to 
the model?

More information about the R-help mailing list