[R-sig-Geo] Collinearity test domain

[Robert J. Hijmans]

Maurizio,
One typically cares about collinearity because it can lead to a sub
optimal model. The model is only affected by the data it sees.
Therefore only the values at your presence/absence points matter.
Robert

On Thu, Sep 4, 2014 at 7:40 AM, Maurizio Marchi
<mauriziomarchi85 at gmail.com> wrote:
> Hallo everybody,
> I have a question about Ecological Niche Modelling / Species Distribution
> Modelling.
>
> If I want to study the distribution of a species in present and future
> time, to remove redundancy and collinearity between predictors should I
> test predictors' correlation/covariance/collinearity using a database
> including only presence/absence available points or should I have to use
> the whole region I'm studing (e.g. the whole Europe?)
>
> as example:
> Presence points: Forest national inventory surveys classified as "pinus
> nigra stands"
> Absence/Pseudo absence points: Forest national inventory surveys NOT
> classified as "pinus nigra stands"
> Study region: Italy
>
> many thanks,
> Maurizio
>
> --
> Maurizio Marchi, Ph.D. student
> Florence, Italy
> ID skype: maurizioxyz
> Ubuntu 14.04 LTS
> linux user 552742
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo