[R] lme: model variance and error by group

Tursiops j.a.balbuena at uv.es
Fri May 21 16:14:08 CEST 2010

I am trying to reproduce a mixed regression model published in (J Evol
Biol 12:94-102) with PASW 17.03. The authors used SAS to fit the model
using the following syntaxis:

proc mixed nobound;
class area individ trait;
model a = side trait area side*trait side*area trait*area/ddfm =
satterthwaite solution;
random intercept side/subject = individ group = areatrai;
repeated/local = quad(areatrai) subject = individ;

The model specifies separate variance components and error variances
for the grouping "areatrai" variable. This is indicated by "group= areatrai"
and "local =
quad (areatrai)", respectively.

Since I currently have no access to SAS and this part of the analysis is
absolutely central to my research, I would be very grateful if someone can
tell whether it is possible to model the variance and error by group using
lme or any other R package for linear mixed models. 

Any help will be much appreciated.


View this message in context: http://r.789695.n4.nabble.com/lme-model-variance-and-error-by-group-tp2226207p2226207.html
Sent from the R help mailing list archive at Nabble.com.

More information about the R-help mailing list