[R-sig-Geo] About extcoeff function.-

[Michael Sumner]

The runif function has arguments min/max:

args(runif)
#function (n, min = 0, max = 1)
#NULL

So, in the first example the sampling is done in [0, 10] and the
second in [100, 200] - so the approximate centre is (5,5) and (150,
150) respectively.

In terms of the extcoeff function, the resulting plot centre is to do
with the modelling process - some transformation that is not simply
related to the actual coordinates of the data.

Someone here may be familiar with this package, but in general terms
the package author is perhaps a more appropriate target for your
questions than this list. There is a vignette in the package, if you
have not already read it:

vignette("SpatialExtremesGuide")

Cheers, Mike.



On Tue, Jun 8, 2010 at 12:12 AM, Freddy López <folopez at ivic.gob.ve> wrote:
> Dear All,
>
> First of all, I present my apologise for this question. I'm sure this
> has an easy solution; but I've no found one yet.
>
> When I run the example of extcoeff function (in SpatialExtremes
> library) I get a plot which axis are latitude and longitude, I
> understand.
>
> In that example the locations are created using:
>
> locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
>
> and I asume that for this reason the 'center' of the plot is about (0,0).
>
> But (and around here emerge my doubts) if I replace the later locations by:
>
> locations <- matrix(runif(2*n.site, 100, 200), ncol = 2)
>
> I could imagine the 'center' of the plot is not (0,0) (perhaps about
> (100,100)?) but this is not the case.
>
> Why this question? Because I have a set of real locations with real
> pluviometrical measures and the plot (extcoeff plot) is refering to
> different coordinates (always around (0,0)).
>
> I don't know if I have misunderstood the help page or something else.
>
> Does anyone know what the problem is?
>
>
> --
> «But Gwindor answered: 'The doom lies in yourself, not in your name.'»
>
> JRR Tolkien
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>