[Biomod-commits] Predictive performance in Ensemble.Forecasting()
Jonathan.Heubes at senckenberg.de
Tue Aug 17 13:56:15 CEST 2010
I've read the Manual carefully. Still, I am not sure about the predictive performance given by the function Ensemble.Forecasting() when the argument Test=TRUE.
Probability values derived from different consensus methods are compared with the data input for a given species using the ROC-curve. But, is the entire data set considered now or the testing data that were kept in memory somewhere? Thanks for an answer. Cheers Jonathan
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Biomod-commits