R help - May 2012 - R crash when i'm using lme function

When I try to adjust a mixed model with random effects I can make this order
without problem
> lm.FA<-lme(absFA~trait*condition,random=~1|individual)
But if I try to fit a model in which the response (absFA) is not the same in
all individuals at different levels of "trait" factor , but varies
randomly
from each. That is, this order
> lm.FA2<-lme(absFA~rasgo*condicion,random=~rasgo|individuo)
R hangs and crashes, I've tried both with R 2.13 and with  R 2.15.

-----
Mario Garrido Escudero
PhD student
Dpto. de Biolog?a Animal, Ecolog?a, Parasitolog?a, Edafolog?a y Qca. Agr?cola
Universidad de Salamanca
--
View this message in context:
http://r.789695.n4.nabble.com/R-crash-when-i-m-using-lme-function-tp4608196.html
Sent from the R help mailing list archive at Nabble.com.

gaiarrido <gaiarrido <at> usal.es> writes:
> When I try to adjust a mixed model with random effects I can make this
order
> without problem
> 
> > lm.FA<-lme(absFA~trait*condition,random=~1|individual)
> 
> But if I try to fit a model in which the response (absFA) is not the same
in
> all individuals at different levels of "trait" factor , but
varies randomly
> from each. That is, this order
> 
> > lm.FA2<-lme(absFA~rasgo*condicion,random=~rasgo|individuo)
> 
> R hangs and crashes, I've tried both with R 2.13 and with  R 2.15.
> 
  This is not reproducible ... and would be better on the
r-sig-mixed-models <at> r-project.org list.  If/when you repost,
more details would be useful:

 * how big is your data set (number of individuals, total
number of observations)?
 * are you including a data= argument in your calls?  It is
generally considered good practice.
 * your predictor variables are different in the two calls --
are you using different data sets, or different predictors,
(i.e. 'condicion' vs 'condition', 'individual' vs
'individuo',
'rasgo' vs 'trait'), or is this a typo?
 * have you checked to see that all variables are of the type
you think they should be (numeric vs factor)?
 * do you actually have multiple observations with different
trait values within at least some individuals?
 * see <http://tinyurl.com/reproducible-000>, or the posting
guide, for information about reproducible examples ...