[GRASSLIST:2144] Re: s.surf.rst and prediction error

[Andrea Aime]

I think you have to use a separate statistical package, such as R, 
to get an estimate of the errors occurred in interpolation (but I'm no
expert
in this field, so I can't tell you how to do it): you'll get this
problem
with every other deterministic interpolation method. As an alternative, 
you can use kriging interpolation method (a geostatistical interpolator)
that provides both interpolated data and an estimate of errors as a
result: 
you can use the latest version of GStat to do it, 
it can read GRASS data (and also output them I suspect...):
www.gstat.org
Hope this helps
Best regards
Andrea Aime

"S . Brewer" wrote:
> 
> Hello All
> 
> On the "Multidimesional interpolation and visualization in GRASS GIS"
> webpage (http://www2.gis.uiuc.edu:2280/modviz/viz/ches.html), there is an
> image showing the predictive error of interpolation.
> 
> I have been using s.surf.rst to interpolate some data, and I would like to
> calculate prediction errors. However, I can't see in the documentation how
> to do this. Is this possible with this module? If so, I would be grateful
> if someone could tell me which syntax to use.
> 
> Thanks in advance
> Simon
> --
> ------------------------------------------------------
> Simon Brewer
> European Pollen Database
> Place de la Republique     tel: 00 33 (0)4 90 96 18 18
> 13200 Arles                fax: 00 33 (0)4 90 93 98 03
> France                  email: simon.brewer at wanadoo.fr
> ------------------------------------------------------