Ben, thanks a lot for your quick reply. Scales of x1 and x2 covariates were too different and this fact may have caused some problem. I have made them similar by dividing one of them by 1000 and now the analysis works. Best regards, Elena
2010/2/1 Ben Bolker <[email protected]> > elena daniela concepcion <elenadconcepcion <at> gmail.com> writes: > > > > > Hi, I'm trying to make multiple poisson regressions with the MLE2 > command. > > I have used the following expression, but I receive an error message: > > > > poisfit <- mle2(y ~ dpois(exp(b0 + b1*x1 + b2*x2)), start=list(b0=1, > b1=1, > > b2=1), data=data1) > > Error in optim(par = c(1, 1, 1), fn = function (p) : > > non-finite initial value 'vmmin' > > > > I have changed initial values using coefficient values given by the GLM > > standard procedure but it still doesn't work. > > Do someone knows where is the problem? > > > > There's nothing obviously wrong here. A reproducible example > would be very helpful ... this *should* give the same result > as glm(y~x1+x2,family=poisson,data=data1) ... What is > with(data1,range(exp(b0+b1*x1+b2*x2))) ? > > ______________________________________________ > [email protected] mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html<http://www.r-project.org/posting-guide.html> > and provide commented, minimal, self-contained, reproducible code. > [[alternative HTML version deleted]] ______________________________________________ [email protected] 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.
