[Biomod-commits] Adding histogram (X=Expl.Var) to response curves

Joseph Maina mainajm at gmail.com
Fri May 18 15:23:22 CEST 2012


Dear Wilfred,

Thanks for the very quick reply; this will do.

Regards,
Joseph

On Fri, May 18, 2012 at 11:01 PM, Wilfried Thuiller
<wilfried.thuiller at ujf-grenoble.fr> wrote:
> Dear Joseph,
> Better to do it sequentially. There is the option to save the response
> curves (similar to the option in box plot, see ?response.plot). That way you
> can really play with the response curves
>
> I have done it the BIOMOD data. Not easy to see anything nice. I would
> rather recommend the rug option instead of the histogram to show the density
> of points
>
> library(BIOMOD)
>
> data(Sp.Env)
>
> #This command is necessary for the run of BIOMOD as a new dataframe is
> produced for the Models function
> Initial.State(Response=Sp.Env[,12:13], Explanatory=Sp.Env[,4:10],
> IndependentResponse=NULL, IndependentExplanatory=NULL)
>
> #Let's run only a few models : GLM, FDA and RF. This will take a few moments
> to run.
> #Do note that the response.plot() function works for all the models
> available. Feel free to switch them on or off.
> Models(GLM = TRUE, TypeGLM = "quad", Test = "AIC", FDA = TRUE, CV.tree = 50,
> RF = TRUE, CV.ann = 2,
>    NbRunEval = 0, DataSplit = 80, Roc=TRUE, Optimized.Threshold.Roc=TRUE,
> Kappa=TRUE, TSS=TRUE, VarImport=5,
>    NbRepPA=0, strategy="sre", coor=NULL, distance=0, nb.absences=1000)
>
> #The response plots for the first species modelled
> load("models/Sp290_GLM_full")
> GLM.RC <- response.plot(Sp290_GLM_full, Sp.Env[,4:10], plot=F,
> save.file="yes")
>
> par(mfrow=c(3,3))
> for(i in 1:ncol(Sp.Env[,4:10])){
> plot(GLM.RC[,1,i], GLM.RC[,2,i],  type="l", ylim=c(0,1),
> xlab=colnames(GLM.RC[,1,])[i], ylab="Probability of occurrence")
> #hist(x=Sp.Env[,3+i],add=T,freq=F)
> rug(Sp.Env[,3+i])
> }
>
>
>
> Hope it helps,
> Wilfried
>
>
>
>
> Le 18 mai 2012 à 12:38, Joseph Maina a écrit :
>
> Hi,
>
> I am trying to add histograms to response curves which were generated
> from a model using the 'response.plot' function. For each variable in
> Expl.Var, histograms should be plotted with the corresponding response
> curve. My script below adds a histogram only to the last response plot
> in the graph. I would appreciate any help with suggestions on how to
> add a histogram to the response plots, where x in the histogram is
> based on Expl.Var or the x-axis of the response curve.
>
> Thanks.
> JOSEPH
>
> response.plot(speciesA_GLM_full, Expl.Var, show.variables =
> seq(1:ncol(Expl.Var)))
> for(i in seq(1:ncol(Expl.Var){
>      hist((x=Expl.Var[,i]),add=TRUE,freq=FALSE)}
>
>
> --------------------------
> Dr. Wilfried Thuiller
> Laboratoire d'Ecologie Alpine, UMR CNRS 5553
> Université Joseph Fourier
> BP53, 38041 Grenoble cedex 9, France
> tel: +33 (0)4 76 51 44 97
> fax: +33 (0)4 76 51 42 79
>
> Email: wilfried.thuiller at ujf-grenoble.fr
> Personal website: http://www.will.chez-alice.fr
> Team website: http://www-leca.ujf-grenoble.fr/equipes/emabio.htm
>
> ERC Starting Grant TEEMBIO
> project: http://www.will.chez-alice.fr/Research.html
> FP6 European EcoChange project: http://www.ecochange-project.eu
>
>
>
>
>
>
>


More information about the Biomod-commits mailing list