It sounds like the function .negloglik.boxcox is being fed some parameters it doesn't like (and possibly it doesn't like the starting values). Possibly attempting to divide by zero. So I would first check .negloglik.boxcox with the starting values to see if something other than infinity comes out.
Jim
-----Original Message-----
From: R-sig-Geo [mailto:r-sig-geo-bounces at r-project.org] On Behalf Of Jimmy Neutr?n
Sent: Tuesday, March 24, 2015 1:46 PM
To: r-sig-geo at r-project.org
Subject: [R-sig-Geo] Maximizing likelihood function
Dear comRades:
I'm working with the variogram "mattern". Then, I tried to run this function many times to achieve the best parameters for my geodata function (XXSelg) by means of likfit, then I wrote:
likfit(XXSelg, cov.model="matern", ini.cov.pars=c(4.0, 1.8), kappa=3.5, fix.kappa=FALSE, nugget=0.8, lambda=0.01, fix.lambda=FALSE, hessian=TRUE) however, many times I received the following answer:
likfit: end of numerical maximisation.Error en optim(par = 1, fn = .negloglik.boxcox, method = "L-BFGS-B", lower = limits$lambda["lower"], : L-BFGS-B needs finite values of 'fn'
One user mentioned that I have to change the method in optim. But likfit only accept RELM or ML.
Then, how can I change the method in likfit???. I wonder if it possible to do it in likfit, because I don't understand the optim script very well (in R-help for package stats).
Does it mean that I can use any of this method = c("Nelder-Mead", "BFGS", "CG", "L-BFGS-B", "SANN"), which are mentioned in optim???
Thanks in advance for youR contribution.
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo