[R] geoR, variofit/likfit

[katiab81]

hi,
I have a problem with the functions variofit ant likfit.
I have to chose the more appropriate variogram model for my specific problem
and data.
I write the alghoritm has follows:

library(geoR)
data(wolfcamp)
v.e<-variog(wolfcamp)
exp.vario<-variofit(v.e,ini.cov.pars=c(175226.94,139.81),cov.model="exponential",nugget=21903.37)
lines(exp.vario,col=1)
sph.vario<-variofit(v.e,ini.cov.pars=c(175226.94,139.81),cov.model="spherical",nugget=21903.37)
lines(sph.vario,col=2)
.....
exp.lik<-likfit(wolfcamp,coords=wolfcamp$coords,data=wolfcamp$data,ini.cov.pars=c(175226.94,139.81),cov.model="exponential",trend="cte",fix.kappa=FALSE,nugget=21903,37)
lines(exp.lik,col=5)
sph.lik<-likfit(wolfcamp,coords=wolfcamp$coords,data=wolfcamp$data,ini.cov.pars=c(175226.94,139.81),cov.model="spherical",trend="cte",fix.kappa=FALSE,nugget=21903,37)
lines(sph.lik,col=6)
....

the lines of the resultant variograms are different using variofit or
likfit, but when I change the cov.model (from exp to sph or lin....) the
program gives identical result.

I think this is incorrect....
can you tell me if I made any mistake in the implementations of the
algorithm. 

thank you. 
P.S. sorry for my english.

katia



--
View this message in context: http://r.789695.n4.nabble.com/geoR-variofit-likfit-tp4013734p4013734.html
Sent from the R help mailing list archive at Nabble.com.