[R] glm cannot find valid starting values

Michael Dewey info at aghmed.fsnet.co.uk
Mon Oct 16 21:34:28 CEST 2006


At 16:12 15/10/2006, Ronaldo ReisJunior wrote:
>Em Sábado 14 Outubro 2006 11:15, Gabor Grothendieck escreveu:
> > Try using OLS starting values:
> >
> > glm(Y~X,family=gaussian(link=log), start = coef(lm(Y~X)))
> >
>
>Ok, using a starting value the model work.
>
>But, sometimes ago it work without any starting values. Why now I need a
>starting values?

Since this involves a prediction about how a 
computer will work in the future I would have 
thought the quote from Samuel Goldwyn in your .sig was helpful.

More seriously I think you need to give more 
detail about when it worked for those more expert 
than me to tell you what happened.


>Thanks
>Ronaldo
>--
>Nunca faça previsões, especialmente sobre o futuro.
>                 -- Samuel Goldwyn
>--
> > Prof. Ronaldo Reis Júnior
>|  .''`. UNIMONTES/Depto. Biologia Geral/Lab. Ecologia Evolutiva
>| : :'  : Campus Universitário Prof. Darcy Ribeiro, Vila Mauricéia
>| `. `'` CP: 126, CEP: 39401-089, Montes Claros - MG - Brasil
>|   `- Fone: (38) 3229-8190 | chrysopa at gmail.com
>| ICQ#: 5692561 | LinuxUser#: 205366

Michael Dewey
http://www.aghmed.fsnet.co.uk



More information about the R-help mailing list