deletions | additions       diff --git a/materials_and_methods.tex b/materials_and_methods.tex  index 41c175c..bc8d3b0 100644  --- a/materials_and_methods.tex  +++ b/materials_and_methods.tex 
...
1-((cases - 1)/(cases - predictors))*(1 - ((null deviance - deviance) / null deviance)).  \item An approximation to the Area Under the Curve (CITATION Fielding and Bell) is computed as the proportion of pseudoabsences with an habitat suitability value lower than the habitat suitability of the test presence.  \end{enumerate}  Finally, all the adjusted explained deviance and AUC values were averaged for each model. We selected all models with AUC values higher than 0.65 and adjusted explained deviance values above 0.1. We computed the average and the standard deviation of the group of  best models (CITATION ENSEMBLE). The average, scaled from 0  to build 1, represented habitat suitability values, while the standard deviation represented the level of agreement reached by the best models. We plotted both measures into a single map, using the \textit{whitening} method proposed by \cite{Hengl200475} to enhance visualization and model interpretation. In  the final ensemble (CITATION ENSEMBLE). model, areas with high habitat suitability values and low standard deviation (good agreement among models) were considered robust indicators of suitable habitat for Neanderthals, while low habitat suitability values plus low standard deviation were considered good indicators of a harsh habitat, and probably, absence.  \textbf{Importance of environmental factors}  Our main goal is to extact as much ecological information as possible from both the data and the model to reach a better understanding of Neanderthals ecology. To do so we analyzed the influence of the environmental factors over the habitat suitability from a continental and a local scale.  We applied Random Forest (CITATION) to analyze the influence of the environmental factors over Neanderthals habitat suitability. The ability of Random Forest to deal with non-linearity makes it perfect to analyse our ensemble, since non-linearity may arise when averaging the results of multiple GLMs.