plot.xvalid, simple question

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

plot.xvalid, simple question

giovanna jona lasinio
When plotting results from a crossvalidation done via xvalid (in geoR) it is
not clear the coloring code used for the coordinate plots. Can anyone
clarify what corresponds to each color? (red and blue in my plots).

Thanks
Giovanna Jona Lasinio



Reply | Threaded
Open this post in threaded view
|

plot.xvalid, simple question

Ole F. Christensen
Hi

Blue corresponds to error being positive and red to error being
negative. The size of the symbols correspond to the absolute value of
the error.

Here ``error''  =  'data - predicted'.

Hope this helps.

Ole

[To Paulo : should  this and the other plots produced by this function
be documented ?]


giovanna jona lasinio wrote:

>When plotting results from a crossvalidation done via xvalid (in geoR) it is
>not clear the coloring code used for the coordinate plots. Can anyone
>clarify what corresponds to each color? (red and blue in my plots).
>
>Thanks
>Giovanna Jona Lasinio
>
>_______________________________________________
>R-sig-Geo mailing list
>R-sig-Geo at stat.math.ethz.ch
>https://www.stat.math.ethz.ch/mailman/listinfo/r-sig-geo
>
>
>  
>

--
Ole F. Christensen
Center for Bioinformatik
Datalogisk Institut
Aarhus Universitet
Ny Munkegade, Bygning 540
8000 Aarhus C
Denmark



Reply | Threaded
Open this post in threaded view
|

R: plot.xvalid, plo.geodata and color coding

giovanna jona lasinio
Thanks for the explanation!

I believe it would be very useful to document at least the color coding of
plots in geoR, colors perception is very subjective, to me red is positive
and blue negative, while in the plot.xvalid the color coding is the
opposite. In the plot.geodata the spatial locations plot is color coded
without a legend, then it's a useful plot which  can be only partially
used...

 Giovanna Jona Lasinio

-----Messaggio originale-----
Da: Ole F. Christensen [mailto:olefc at daimi.au.dk]
Inviato: Thursday, November 06, 2003 6:47 PM
A: giovanna jona lasinio
Cc: r-sig-geo at stat.math.ethz.ch; Paulo Ribeiro
Oggetto: Re: [R-sig-Geo] plot.xvalid, simple question


Hi

Blue corresponds to error being positive and red to error being
negative. The size of the symbols correspond to the absolute value of
the error.

Here ``error''  =  'data - predicted'.

Hope this helps.

Ole

[To Paulo : should  this and the other plots produced by this function
be documented ?]


giovanna jona lasinio wrote:

>When plotting results from a crossvalidation done via xvalid (in geoR) it
is

>not clear the coloring code used for the coordinate plots. Can anyone
>clarify what corresponds to each color? (red and blue in my plots).
>
>Thanks
>Giovanna Jona Lasinio
>
>_______________________________________________
>R-sig-Geo mailing list
>R-sig-Geo at stat.math.ethz.ch
>https://www.stat.math.ethz.ch/mailman/listinfo/r-sig-geo
>
>
>
>

--
Ole F. Christensen
Center for Bioinformatik
Datalogisk Institut
Aarhus Universitet
Ny Munkegade, Bygning 540
8000 Aarhus C
Denmark