[R] nlme : convergence problem and other errors

From: Thomas BRUNEL <Thomas.Brunel_at_ifremer.fr>
Date: Tue 16 Jan 2007 - 13:48:19 GMT


Dear R-user,

I am trying to use the R "nlme" function to fit a non linear mixed effects model. The model I wand to fit is an individual somatic growth model with 4 parameters. For all parameters both fixed and random effects have to be estimated, as well as their covariance matrix (see the formula bellow).
The data are simulated with the same growth model as in the nlme, with know parameters, and covariance matrix.

I tried to fit the model with several simulated data sets, but most of the time, R returns an error message.
there are three main types of errors :

- there is a singular matrix
- a factor is reduced bellow the PNLS level.
- max number of iteration is reached but there is no convergence.


Do you know how to resolve these problems. Is there a way to modify the parameters of the maximization algorithm to avoid these error messages?

Furthermore, do you know if it is possible to fix the values of the fixed effects so that the model only has to estimate the random effects?

Thank you for your help and answers.

Regards,

Thomas Brunel

pds.fit<-nlme(pds~(exp (lna)/(exp (lnb) + 1/(1 + exp(1000 * (exp
(lntmat) - t + 0.5))) * exp (lnc)))^4 * (1 - (1 - ( 0.051 * (1 - 1/(1 +
exp(1000 * (exp (lntmat) - t + 0.5)))) + (exp (lna)/exp (lnb))^4 * (1 -
(1 - (0.051)^0.25 * (exp (lnb)/exp (lna))) * exp( - ((exp (lnb) * exp
(lntmat))/4)))^4 * 1/(1 + exp(1000 * (exp (lntmat) - t + 0.5))))^0.25 *
((exp (lnb) + 1/(1 + exp(1000 * (exp (lntmat) - t + 0.5))) * exp
(lnc))/exp (lna))) * exp( - (((exp (lnb) + 1/(1 + exp(1000 * (exp
(lntmat) - t + 0.5))) * exp (lnc)) *(t - (1/(1 + exp(1000 * (exp
(lntmat) - t + 0.5))) * exp
(lntmat))))/4)))^4,data=pdsdata,fixed=lna+lnb+lnc+lntmat~1,random=
lna+lnb+lnc+lntmat~1|indi ,start=c(lna=log(a2),lnb=log(b2) ,lnc = log(c2), lntmat=log(1200)))

-- 


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Laboratoire Halieutique de Port-en-Bessin


avenue du Général de Gaulle
14520 Port-en-Bessin
tél : 02 31 51 56 00 (standard)
fax : 02 31 51 56 01


page personnelle (CV online):
http://www.ifremer.fr/drvrhbr/personnel/brunel/index.htm

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Received on Wed Jan 17 01:10:41 2007

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Tue 16 Jan 2007 - 14:30:27 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.