[R-sig-Geo] geoR likfit message

[Paulo Justiniano Ribeiro Jr]

Not really Terry

tha kappa parameter is an "extra"parameters used in some but not all
models.
Spherical end exponential models for the correlation functions are cases
where kappa is note used and, therfore, the argument makes no diference

best
P.J.


Paulo Justiniano Ribeiro Jr
LEG (Laboratório de Estatística e Geoinformação)
Universidade Federal do Paraná
Caixa Postal 19.081
CEP 81.531-990
Curitiba, PR  -  Brasil
Tel: (+55) 41 3361 3573
Fax: (+55) 41 3361 3141
e-mail: paulojus AT  ufpr  br
http://www.leg.ufpr.br/~paulojus

On Thu, 30 Aug 2007, Terry Griffin wrote:

> Greetings,
>
> I am revisiting some simulations that I ran about a year ago.  When I run the likfit function from geoR; a message is returned that reads "kappa not used for the spherical correlation function".  The simulations still run, but wasn't sure if this message indicates another problem that I need to be aware of.  I'm not using "fix.kappa=TRUE" or other references to kappa.  Below is the line in the code that is returning the message.  The "cressie" refers to the variofit function.
>
> geo.reml<-likfit(data.geo,coords=coords.geo,data=y,cov.model="spherical",trend=~s2+s3+s4+t+ts2+ts3+ts4,method.lik="REML",ini=c(cressie[2]$cov.pars[1],cressie[2]$cov.pars[2]), messages=FALSE)#
>
> Is this anything to be concerned about?
>
> Thank you,
>
> Terry
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>