This function evaluates and averages simple bivariate models by weighted means to Ensemble Small Models as in Lomba et al. 2010 and Breiner et al. 2015.
ecospat.ESM.EnsembleModeling( ESM.modeling.output,
weighting.score,
threshold=NULL,
models)
species: species name ESM.fit: data.frame of the predicted values for the data used to build the models. ESM.evaluations: data.frame with evaluations scores for the ESMs weights: weighting scores used to weight the bivariate models to build the single ESM weights.EF: weighting scores used to weight the single ESM to build the ensemble of ESMs from different modelling techniques (only available if >1 modelling techniques were selected). failed: bivariate models which failed because they could not be calibrated.
A BIOMOD.ensemble.models.out-class object. This object will be later given to ecospat.ESM.EnsembleProjection if you want to make some projections of this ensemble-models.
a list returned by ecospat.ESM.Modeling
an evaluation score used to weight single models to build ensembles:"AUC","TSS",
"Boyce","Kappa","SomersD"
( see BIOMOD_Modeling, metric.eval section for more detailed informations )
threshold value of an evaluation score to select the bivariate model(s) included for building the ESMs
vector of models names choosen among 'GLM', 'GBM', 'GAM', 'CTA', 'ANN', 'SRE', 'FDA', 'MARS', 'RF','MAXENT', "MAXNET" (same as in biomod2)
#a character vector (either 'all' or a sub-selection of model names) that defines the models kept for building the ensemble models (might be useful for removing some non-preferred models)
Frank Breiner frank.breiner@wsl.ch
with contributions of Olivier Broennimann and Flavien Collart olivier.broennimann@unil.ch
The basic idea of ensemble of small models (ESMs) is to model a species distribution based on small, simple models, for example all possible bivariate models (i.e. models that contain only two predictors at a time out of a larger set of predictors), and then combine all possible bivariate models into an ensemble (Lomba et al. 2010; Breiner et al. 2015).
The ESM set of functions could be used to build ESMs using simple bivariate models which are averaged using weights based on model performances (e.g. AUC) according to Breiner et al. (2015). They provide full functionality of the approach described in Breiner et al. (2015).
Lomba, A., L. Pellissier, C.F. Randin, J. Vicente, F. Moreira, J. Honrado and A. Guisan. 2010. Overcoming the rare species modelling paradox: A novel hierarchical framework applied to an Iberian endemic plant. Biological Conservation, 143,2647-2657.
Breiner F.T., A. Guisan, A. Bergamini and M.P. Nobis. 2015. Overcoming limitations of modelling rare species by using ensembles of small models. Methods in Ecology and Evolution, 6,1210-1218.
Breiner F.T., Nobis M.P., Bergamini A., Guisan A. 2018. Optimizing ensembles of small models for predicting the distribution of species with few occurrences. Methods in Ecology and Evolution. tools:::Rd_expr_doi("10.1111/2041-210X.12957")
ecospat.ESM.Modeling