[R-sig-Geo] variograms do not satisfy a legal model

[Edzer J. Pebesma]

Thierry, you'll have to find out what the linear model of 
regionalization or intrinsic correlation mean; they are legal models, 
meaning that they lead to predictions with guaranteed non-negative 
prediction variances. There's a way to override the check (meaning it 
will not stop on error) but not to get that guarantee.
--
Edzer

ONKELINX, Thierry wrote:
> Dear useRs,
>
> I'm trying to do some cokriging. I try fit the variograms with fit.lmc
> (gstat package).
>
> #Problem 1: how to get the model in an elegant way into the gstat
> object? This is how I do it now.
> library("gstat")
> data(meuse)
> g <- gstat(id = "ln.zinc", formula = log(zinc)~1, locations = ~x+y, data
> = meuse)
> g <- gstat(g, id = "ln.lead", formula = log(lead)~1, locations = ~x+y,
> data = meuse)
> #fit the variogram models using fit.lmc
> model <- fit.lmc(variogram(g), g, model = vgm(.55, "Sph", 900, .05),
> fit.ranges = FALSE)
> g <- gstat(g, id = "ln.zinc", model = model[["model"]][["ln.zinc"]])
> g <- gstat(g, id = "ln.lead", model = model[["model"]][["ln.lead"]])
> g <- gstat(g, id = c("ln.zinc", "ln.lead"), model =
> model[["model"]][["ln.zinc.ln.lead"]])
> predict(g, newdata = meuse)
>
> #Problem 2. the gstat object doesn't accept different ranges. Or am I
> doing something wrong?
> library("gstat")
> data(meuse)
> g <- gstat(id = "ln.zinc", formula = log(zinc)~1, locations = ~x+y, data
> = meuse)
> g <- gstat(g, id = "ln.lead", formula = log(lead)~1, locations = ~x+y,
> data = meuse)
> # examine variograms and cross variogram:
> plot(variogram(g))
> # enter direct variograms:
> g <- gstat(g, id = "ln.zinc", model = vgm(.55, "Sph", 800, .05))
> g <- gstat(g, id = "ln.lead", model = vgm(.55, "Sph", 900, .05))
> # enter cross variogram:
> g <- gstat(g, id = c("ln.zinc", "ln.lead"), model = vgm(.47, "Sph", 900,
> .03))
> predict(g, newdata = meuse)
> # Error in predict.gstat(g, newdata = meuse) : gstat: value not allowed
> for: variograms do not satisfy a legal model
>
> Thanks,
>
> Thierry
>
> ------------------------------------------------------------------------
> ----
> ir. Thierry Onkelinx
> Instituut voor natuur- en bosonderzoek / Reseach Institute for Nature
> and Forest
> Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
> methodology and quality assurance
> Gaverstraat 4
> 9500 Geraardsbergen
> Belgium
> tel. + 32 54/436 185
> Thierry.Onkelinx at inbo.be
> www.inbo.be 
>
> Do not put your faith in what statistics say until you have carefully
> considered what they do not say.  ~William W. Watt
> A statistical analysis, properly conducted, is a delicate dissection of
> uncertainties, a surgery of suppositions. ~M.J.Moroney
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>