Hello everybody!
I'm having some trouble to compute maximum likelihood estimations using maxLik
package and I hope that you could give me a hint.
The main problem is that I'm not able to get a result not even close to the
ones given by glm() directly, and the second one is: "Error in maxNRCompute(fn
= logLikAttr, fnOrig = fn, gradOrig = grad, hessOrig = hess, : NA in gradient".
The codes:
loglike.GGompiMaxLik <- function(theta,age,deaths,exposures) {
alpha <- exp(theta[1])
beta <- exp(theta[2])
gamma <- exp(theta[3])
first <- alpha*exp(beta*age)
second <- 1+(((alpha*gamma)/beta)*(exp(beta*age)-1))
mu <- first/second
llk <- -sum((deaths * log(mu)) + (- mu*exposures))
return(llk)
}
fit1 <- maxLik(loglike.GGompiMaxLik,
age=0:6,
deaths=c(15545, 21278, 32444, 36201, 30360, 14201, 5198),
exposures=c(935023.67, 819392.00, 724568.17, 470947.00,
231951.64, 69502.65, 15798.72),
start=c(-4.1402817, -0.6375773, -1.6945914))
Do you know how I can solve this problem?
Thanks in advance,
Filipe
[[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.
