[R-sig-Geo] gstat fit.variogram, attr(fit.1,"SSErr")

[Paul Hiemstra]

On 08/23/2010 05:45 PM, Kerry Ritter wrote:
> Hi. I am using library "gstat" in R.  My question is why do I get 
> different SSErr's when I fit the variogram model using fit.variogram 
> than when I fix the parameters in fit.variogram using the exact same 
> parameters obtained from the first fit.  My expectation is that these 
> would be the same.  What is SSErr actually calculating?
> Thanks,
> Kerry
>

Hi Kerry,

The following example illustrates that fit.variogram is sensitive to the 
start parameters you use for Gauss-Newton fitting:

data(meuse)
vgm1 <- variogram(log(zinc)~1, ~x+y, meuse)
v1 = fit.variogram(vgm1, vgm(1,"Sph",300,1))

# Use fit above as initial guess
v2 = fit.variogram(vgm1, v1)

v1
#  model     psill    range
#1   Nug 0.0506555   0.0000
#2   Sph 0.5906009 896.9702

v2
#   model      psill   range
# 1   Nug 0.05066039   0.000
# 2   Sph 0.59060578 897.006

So the difference in SSerr is caused by a different fit by 
fit.variogram, which in turn is caused by a different initial variogram 
model guess.

regards,
Paul

-- 
Drs. Paul Hiemstra
Department of Physical Geography
Faculty of Geosciences
University of Utrecht
Heidelberglaan 2
P.O. Box 80.115
3508 TC Utrecht
Phone:  +3130 253 5773
http://intamap.geo.uu.nl/~paul
http://nl.linkedin.com/pub/paul-hiemstra/20/30b/770