[R-sig-Geo] Collinearity test

[b.naimi at utwente.nl]

Hi Maurizio,

You can use vif, vifstep, vifcor and exclude functions in usdm package to deal with collinearity problem in environmental variables. The input variables can be either raster object or data.frames. You can find an example of using these function for SDM in the following reference:

Naimi, B.,et al..(2013). Where is positional uncertainty a problem for species distribution modelling, Ecography. DOI: 10.1111/j.1600-0587.2013.00205.x<http://onlinelibrary.wiley.com/doi/10.1111/j.1600-0587.2013.00205.x/abstract>

Hope this helps.

Cheers,

Babak
http://r-gis.net


> -----Original Message-----
Date: Sun, 17 Nov 2013 12:03:18 -0800
From: "Robert J. Hijmans" <r.hijmans at gmail.com>
To: "Proosdij, Andre van" <andre.vanproosdij at wur.nl>
Cc: "r-sig-geo at r-project.org" <r-sig-geo at r-project.org>,        Maurizio
        Marchi <mauriziomarchi85 at gmail.com>
Subject: Re: [R-sig-Geo] Collinearity test
Message-ID:
        <CANtt_hw=3-EtHj+B3AjowNTX3_7ZDhGUWexDa3OzVWFRjCEOeA at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

Also see:

library(dismo)
s <- stack(system.file("external/rlogo.grd", package="raster"))
pairs(s)

Robert

On Wed, Nov 13, 2013 at 3:06 AM, Proosdij, Andre van
<andre.vanproosdij at wur.nl> wrote:
> Hi Maurizio,
>
> I used the rcorr function from the Hmisc package to calculate the Pearson correlation.
>
> Best,
>
>
> Andr?
>
>
> Ir. A.S.J. van Proosdij
> PhD Student
> e  andre.vanproosdij at wur.nl
> Naturalis Biodiversity Center, section NHN
> Biosystematics Group, Wageningen University
> Gen. Foulkesweg 37, 6703 BL Wageningen, the Netherlands
> www.bis.wur.nl<http://www.bis.wur.nl>
>

> -----Original Message-----
> From: r-sig-geo-bounces at r-project.org [mailto:r-sig-geo-bounces at r-project.org] On Behalf Of Maurizio Marchi
> Sent: woensdag 13 november 2013 11:00
> To: r-sig-geo at r-project.org
> Subject: [R-sig-Geo] Collinearity test
>
> Hallo everybody,
> I'm working with Species Distibution Models to predict the potential distribution of a forest specie.
> I use WorldClim raster data for Italy and before running SDMs algorithm I would like to test the collineariy between factors to eliminate redundancy of data.
> My question in: is there a package that can be used to Test collinearity between factors of a raster stack (worldclim biovariables)?
>
> Thanks,
>
> --
> Maurizio Marchi, Ph.D student
> Forestry Research Centre (Arezzo, IT)
> ID skype: maurizioxyz
> *Ubuntu 12.04 (Precise Pangolin)*
>
>         [[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
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo

        [[alternative HTML version deleted]]