<div dir="ltr"><div>This is the the data set and want to make variogram for Cd15. The distance between each sample location was 4 cross 4 km. Please suggest me how to allocate initial guess about nugget, sill and range parametrs. Moreover, I have used log transformation for my variable Cd15. Is it appropriate? Or there is any other transformation for this.</div><div>Thanks,</div><div>Regards</div><div>Adeela</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Jun 24, 2015 at 7:08 PM, Edzer Pebesma <span dir="ltr"><<a href="mailto:edzer.pebesma@uni-muenster.de" target="_blank">edzer.pebesma@uni-muenster.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Thierry is right about that; without that we can only guess, which is<br>
what I will do now.<br>
<br>
fit.variogram does a non-linear fit using some steepest descent method,<br>
and needs initial values for that. The initial values you provide are 1<br>
(partial sill), 5 (range), and 1 (nugget). In case v does not provide<br>
good information on how to improve these initial values, e.g. because 5<br>
is smaller than the smallest distance in it, it can't figure out where<br>
to go in order to improve. This means it got stuck, and gives this<br>
warning. The plot<br>
<br>
plot(v, vgm(1, "Sph", 5, 1))<br>
<br>
will reveal the problem to you (and to us, if you'd share it).<br>
<div class="HOEnZb"><div class="h5"><br>
On 06/24/2015 03:42 PM, Thierry Onkelinx wrote:<br>
> You'll need to provide a reproducible example. Your code is not<br>
> reproducible without the data.<br>
><br>
> ir. Thierry Onkelinx<br>
> Instituut voor natuur- en bosonderzoek / Research Institute for Nature and<br>
> Forest<br>
> team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance<br>
> Kliniekstraat 25<br>
> 1070 Anderlecht<br>
> Belgium<br>
><br>
> To call in the statistician after the experiment is done may be no more<br>
> than asking him to perform a post-mortem examination: he may be able to say<br>
> what the experiment died of. ~ Sir Ronald Aylmer Fisher<br>
> The plural of anecdote is not data. ~ Roger Brinner<br>
> The combination of some data and an aching desire for an answer does not<br>
> ensure that a reasonable answer can be extracted from a given body of data.<br>
> ~ John Tukey<br>
><br>
> 2015-06-24 15:23 GMT+02:00 adeela uaf <<a href="mailto:adeela.uaf@gmail.com">adeela.uaf@gmail.com</a>>:<br>
><br>
>> Hi,<br>
>> Can anybody help me in fitting variogram in gstat package in R. I am trying<br>
>> but having this problem. Where Cd15 is the amount of Cadmium in soil at 15<br>
>> cm depth.<br>
>> v<-variogram(Cd15~1,data)<br>
>> v.fit<-fit.variogram(v,vgm(1,"Sph",5,1))<br>
>> Warning message:<br>
>> In fit.variogram(v, vgm(1, "Sph", 5, 1)) :<br>
>> Warning: singular model in variogram fit<br>
>> ******************************<br>
>> Adeela<br>
>> University of Agriculture Faisalabad<br>
>> Department of Math & Stat.<br>
>><br>
>> [[alternative HTML version deleted]]<br>
>><br>
>> _______________________________________________<br>
>> R-sig-Geo mailing list<br>
>> <a href="mailto:R-sig-Geo@r-project.org">R-sig-Geo@r-project.org</a><br>
>> <a href="https://stat.ethz.ch/mailman/listinfo/r-sig-geo" target="_blank" rel="noreferrer">https://stat.ethz.ch/mailman/listinfo/r-sig-geo</a><br>
>><br>
><br>
> [[alternative HTML version deleted]]<br>
><br>
> _______________________________________________<br>
> R-sig-Geo mailing list<br>
> <a href="mailto:R-sig-Geo@r-project.org">R-sig-Geo@r-project.org</a><br>
> <a href="https://stat.ethz.ch/mailman/listinfo/r-sig-geo" target="_blank" rel="noreferrer">https://stat.ethz.ch/mailman/listinfo/r-sig-geo</a><br>
><br>
<br>
</div></div><span class="HOEnZb"><font color="#888888">--<br>
Edzer Pebesma<br>
Institute for Geoinformatics (ifgi), University of Münster,<br>
Heisenbergstraße 2, 48149 Münster, Germany; <a href="tel:%2B49%20251%2083%2033081" value="+492518333081">+49 251 83 33081</a><br>
Journal of Statistical Software: <a href="http://www.jstatsoft.org/" target="_blank" rel="noreferrer">http://www.jstatsoft.org/</a><br>
Computers & Geosciences: <a href="http://elsevier.com/locate/cageo/" target="_blank" rel="noreferrer">http://elsevier.com/locate/cageo/</a><br>
Spatial Statistics Society <a href="http://www.spatialstatistics.info" target="_blank" rel="noreferrer">http://www.spatialstatistics.info</a><br>
<br>
</font></span><br>_______________________________________________<br>
R-sig-Geo mailing list<br>
<a href="mailto:R-sig-Geo@r-project.org">R-sig-Geo@r-project.org</a><br>
<a href="https://stat.ethz.ch/mailman/listinfo/r-sig-geo" target="_blank" rel="noreferrer">https://stat.ethz.ch/mailman/listinfo/r-sig-geo</a><br>
<br></blockquote></div><br></div>