[Gstat-info] Non-Interactive curve fitting and prediction

[Edzer J. Pebesma]

Mach Nife wrote:

>Hi,
>
>I've used this command file:
>data(density): 'density.txt', x=2, y=3, v=1, min=4,
>max=40, radius=1000; 
>variogram(density): 'density.sem': 0.14 Nug(0) + 300
>Exp(10);
>method: semivariogram;
>set fit=7;
>
>And this command:
>gstat -l cmd_c.log cmd_c.gst
>
>The fitted formula is shown in the cmd_c.log file as :
>424437 Nug(0) - 87.4479 Exp(-12.784)
>
>Is there a way to modify the above command file, so it
>takes this fitted formula to make the predictions? It
>would be nice if it could all be done in one operation
>(Fit the formula and use this formula right away to
>make the predictions). I've tried adding this at the
>bottom of the command file, without success:
>predictions(density): 'Dens_Krig';
>
>Is it possible?
>  
>
No; you'd need to do some scripting to pull the
fitted model (which is in a syntax gstat can read)
out of the log file and put it into the next command file
with predictions.

In R this would all be fairly easy.
--
Edzer

>machnife
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around 
>http://mail.yahoo.com 
>_______________________________________________
>Gstat-info mailing list
>Gstat-info at geo.uu.nl
>http://mailman.geo.uu.nl/mailman/listinfo/gstat-info
>  
>