[R-sig-Geo] geoR

Ole F. Christensen olefc at daimi.au.dk
Mon Mar 7 08:12:56 CET 2005

Yes, the ``range parameter'' phi is a free parameter.
Calculation of variance of the estimate for phi  is not implemented in 
function likfit.
Instead you should use the proflik function to obtain the profile 
likelihood for this parameter.


Ruben Roa wrote:

>-----Original Message-----
>From:	r-sig-geo-bounces at stat.math.ethz.ch on behalf of Ole F. Christensen
>Sent:	Fri 3/4/2005 3:53 PM
>To:	Anders Malmberg
>Cc:	R-sig-Geo at stat.math.ethz.ch
>Subject:	Re: [R-sig-Geo] geoR
>Dear Anders
>* The maximum likelihood estimate of the partial sill parameter sigmasq 
>can be expressed explictly as function of the other covariance parameters.
>Therefore parameter estimate for sigmasq is NOT found by numerical 
>maximisation, and as consequence in likfit you cannot specify that that 
>sigmasq should be within certain limits.
>* The parameter estimate for the relative nugget tausq.rel is found by 
>numerical maximisation. Therefore you can specify that this parameter 
>should be within certain limits.
>I understand that the range parameter phi is also a 'free parameter'. Am i correct?
>Another quick question: i haven't been able to obtain the estimation variance of the phi parameter estimate as i do for the beta estimate with likfitobject$beta.var. What is the necessary command?
>[snip Anders Malmberg's original question]
>R-sig-Geo mailing list
>R-sig-Geo at stat.math.ethz.ch

Ole F. Christensen
BiRC - Bioinformatics Research Center
University of Aarhus

More information about the R-sig-Geo mailing list