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