[R] spatial: surf.gls() rank failure; was (no subject)

[Roger Bivand]

Please always use an informative subject.

Please also recall that questions about contributed packages should first 
be addressed to their maintainers.

In the case of the spatial package, another source of information that 
should be consulted before contacting the maintainer, is the book that the 
software is wriiten to support.

Finally, consider that a prime benefit of open source software is that you 
can read the code, and identify where error messages are being generated.

On Thu, 22 May 2003, Branimir K. Hackenberger wrote:

> Dear R-helpers!
> 
> What it's means "Rank failure in Choleski decomposition" by using of
> function surf.gls {spatial}?
> 
In this case, reading the code of surf.gls shows that this message is 
displayed when the C function VR_gls() returns a non-zero value of ifail, 
which happens when cholcov() returns from chols(), the algorithm of which 
is given in: M.J.R. Healy (1968) Applied Statistics  pp. 195-197.

If that is too abstract, try:

data(topo)
topo <- rbind(topo, topo[50:52,])
topo.kr <- surf.gls(2, expcov, topo, d = 0.7)
Error in surf.gls(2, expcov, topo, d = 0.7) : 
        Rank failure in Choleski decomposition

that is, you seem to have duplicate point locations, leading to a 
covariance matrix not of full rank.  

> Sincerely Yours
> 
> Branimir K. Hackenberger
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> 

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: Roger.Bivand at nhh.no