R help - Nov 2008 - how to join these two models?

Dear R users,

I have this 2 models that fit to my data:


M3varI <- update (M3, weights=varIdent(form= ~ 1|SITE))

M3AR1<-update(M3,correlation=corAR1())

The first one, updates my M3 so that I can account for the variance  
structure of random erros.
The second one, updates my M3 so that I can account for the  
correlation structure of random errors.


How can I put them toghether in one final model?

I have tryed it using gls but I could not do it.

Could you please help me o this?

Bets regards,

Sara Maltez Mouro

Centro de Ecologia Funcional
Departamento de Botânica
Universidade de Coimbra

sara@gmesintra.com
www.uc.pt/ecology/saramaltezmouro


	[[alternative HTML version deleted]]

Sara Mouro wrote:
> 
> I have this 2 models that fit to my data:
> 
> M3varI <- update (M3, weights=varIdent(form= ~ 1|SITE))
> 
> M3AR1<-update(M3,correlation=corAR1())
> 
> How can I put them toghether in one final model?
> 
> 
This looks like lme in package nlme, but it is pure guesswork, since your
example is incomplete. Probably you should not use update, but something
like

   lme(xxxxx~xxxx, random =....., weights=varIdent..., correlation=corAR1())

but I am not sure if this somewhat overloaded fit will converge.

Dieter


-- 
View this message in context:
http://www.nabble.com/how-to-join-these-two-models--tp20516771p20516980.html
Sent from the R help mailing list archive at Nabble.com.