An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20121217/5acc6fd9/attachment.pl>
How to make Ordinary Kriging using gstat predict?
5 messages · S Ellison, DIMITRIS KARAKOSTIS, Jon Olav Skoien
-----Original Message----- My problem is that instead of Ordinary kriging, when I run the algorithm I get: Inverse distance weighted interpolation. Why is that? What am I missing or doing wrong?
The gstat manual at http://www.gstat.org/gstat.pdf says on p16 that "When no variograms are specified, inverse distance weighted interpolation is the default action (Fig. 2.1, example [6.3]). When variograms are specified the default prediction method is ordinary kriging Journel and Huijbregts (1978); Cressie (1993) (example [6.4] and example [6.8])." It looks like reading that manual may be useful ... S Ellison ******************************************************************* This email and any attachments are confidential. Any use...{{dropped:8}}
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20121217/42db00a4/attachment.pl>
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20121217/c83064be/attachment.pl>
1 day later
Hi Dimitris, The mistake is that predict.gstat doesnt have a "model" argument, as you assume. But as the function also accepts arguments through ..., it does not complain about the unused argument. Try instead to put the model argument in the gstat-object as you can see in the example in ?predict.gstat: g <- gstat(id="tec", formula=TEC ~ 1, data=data, model = v.fit) Cheers, Jon BTW, you will generally get quicker response to questions regarding any kind of spatial data handling from the mailinglist r-sig-geo at r-project.org.
On 17-Dec-12 18:58, DIMITRIS KARAKOSTIS wrote:
Thanks for the answer. I have already read the gstat manual and I had constructed the empirical and theoretical variogram like this: g <- gstat(id="tec", formula=TEC ~ 1, data=data)v <- variogram(g)mod<-vgm(sill=var(data$TEC),model="Sph",range=200,nugget=10)v.fit <- fit.variogram(v, model=mod,fit.method=1)Theor_variogram=plot(variogram(TEC~1,data),v.fit,main="WLS Model")plot(Theor_variogram) But still, when I use predict:p <- predict.gstat(g, model=v.fit, newdata=predGrid) ..instead of ordinary kriging I get inverse distance weighted.Please, if anyone knows where I make the mistake or what I miss, please let me know!Thanks
From: S.Ellison at LGCGroup.com To: dimitriskarakostis3 at hotmail.com; r-help at r-project.org Date: Mon, 17 Dec 2012 17:22:12 +0000 Subject: RE: [R] How to make Ordinary Kriging using gstat predict?
-----Original Message----- My problem is that instead of Ordinary kriging, when I run the algorithm I get: Inverse distance weighted interpolation. Why is that? What am I missing or doing wrong?
The gstat manual at http://www.gstat.org/gstat.pdf says on p16 that "When no variograms are specified, inverse distance weighted interpolation is the default action (Fig. 2.1, example [6.3]). When variograms are specified the default prediction method is ordinary kriging Journel and Huijbregts (1978); Cressie (1993) (example [6.4] and example [6.8])." It looks like reading that manual may be useful ... S Ellison ******************************************************************* This email and any attachments are confidential. Any u...{{dropped:11}}
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Jon Olav Sk?ien Joint Research Centre - European Commission Institute for Environment and Sustainability (IES) Land Resource Management Unit Disclaimer: Views expressed in this email are those of the individual and do not necessarily represent official views of the European Commission.