[Biomod-commits] evaluating results

[Wilfried Thuiller]

Hi Pep,
of course this is possible to have a better value with an ensemble  
forecasting than with the best model. This is even why we use ensemble  
forecasting!

The advantage is that each model bring its own information and that  
all the information put together is more informative than the one  
coming from the best single model.

Hope I am clear.

Wilfried



Le 4 nov. 09 à 17:02, Pep Serra a écrit :

> Sorry, I meant Evaluation.results.TSS when I wrote  
> Evaluation.results.Kappa
>
>
>
>
> En/na Pep Serra ha escrit:
>>
>> Hi BIOMODers,
>>
>> I run BIOMOD for some species and I obtained these results which  
>> are a bit difficult to interpret to me:
>>
>> I am checking AUC and TSS model by model and comparing it to the  
>> Ensemble modeling approach:
>>
>>
>>
>> TSS
>>
>> mean
>> ANN	0.70566667
>> CTA	0.7495
>> GAM	0.75466667
>> GBM	0.77866667
>> GLM	0.7365
>> MARS	0.75016667
>> MDA	0.69866667
>> RF	0.80516667
>> SRE	0.436
>> EF	0.97055927
>> I am producing an ensemble forecasting giving weights according to  
>> the TSS statistic (decay=1.6). For each model I used the cross- 
>> validation result out of Evaluation.results.Kappa.
>> Is it possible to obtain a TSS statistic even much better than the  
>> best model and the best run of the best model (according to TSS)?
>>
>> Thanx !
>> _______________________________________________
>> Biomod-commits mailing list
>> Biomod-commits at lists.r-forge.r-project.org
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/biomod-commits
>>
>
> <josep_serra.vcf>_______________________________________________
> Biomod-commits mailing list
> Biomod-commits at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/biomod-commits

-------------------------
Dr. Wilfried Thuiller
Laboratoire d'Ecologie Alpine, UMR CNRS 5553
Université Joseph Fourier
BP53, 38041 Grenoble cedex 9, France
tel: +33 (0)4 76 63 54 53
fax: +33 (0)4 76 51 42 79

Email: wilfried.thuiller at ujf-grenoble.fr
Home page: http://www.will.chez-alice.fr
Website: http://www-leca.ujf-grenoble.fr/equipes/tde.htm

FP6 European MACIS project: http://www.macis-project.net
FP6 European EcoChange project: http://www.ecochange-project.eu





-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20091104/ca188776/attachment.htm