Dear All:
I tried to fit negative binomial distribution to data in terms of mean and
mean is also a quadratic function of another variable. The likelihood function
is:
function (parameters, y1,x11)
{
p<-parameters[1]
alpha1<-parameters[1]
beta1<-parameters[2]
delta1<-parameters[3]
mu<-alpha1+beta1*(x11)+delta1*(x11^2)
ifelse(y1>=0|x11>=0,
L<-
lgamma(y1+p)+p*(log(p)-log(mu+p))+y1*(log(mu)-log(mu+p))
-lfactorial(y1)-lgamma(p)
,Inf)
L
}
I got outputs even it sounds a problem for delata. Actually, there is a mistake
made for the parameters (p & alpha1) set up. If I correct them as:
p<-parameters[1]
alpha1<-parameters[2]
beta1<-parameters[3]
delta1<-parameters[4]
Then error messga there: initial value in 'vmmin' is not finite
In addition: There were 38 warnings (use warnings() to see them).
Could you give some advice please?
Thanks a lot!
Xin Shi
[[alternative HTML version deleted]]
Xin <jasonshi510 <at> hotmail.com> writes:> > Dear All: >> Then error messga there: initial value in 'vmmin' is not finite > In addition: There were 38 warnings (use warnings() to see them). > > Could you give some advice please? > > Thanks a lot! > > Xin Shi > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help <at> 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 > >A little more information (and an informative Subject: line) would help. A reproducible example (let us know how you're calling optim(), and ideally show your data or a subsample of it that produces the same problem --- see the posting guide). A few ideas: - the 'vmmin' error message says that your starting values are giving non-finite results -- this could be because some of your x and y values are negative (which in your could would lead to infinite values). Bottom line: try your objective function with your starting parameter values and see what happens - you could use dnbinom(...,log=TRUE) if you wanted - you may need to be careful with constraining alpha, beta, delta, as you could end up with negative means.