Publications by A.M. Barbosa
Improvements to fuzzySim functions
A few improvements were recently made to several functions in the fuzzySim package. Mainly, function modelTrim is now more independent (it used to require “attach” sometimes); multTSA allows either “AIC” or “significance” as a backward stepwise selection criterion, and provides more descriptive polynomial term names (visible when save...
1119 sym
Package fuzzySim now on CRAN!
The fuzzySim package, which is regularly maintained on R-Forge, is finally also available on the CRAN official repository, in its latest version 2.0! Many thanks especially to Alba Estrada for invaluable help with bug search. Related To leave a comment for the author, please follow the link and comment on their blog: modTools. R-bloggers.com ...
616 sym
Calculate biotic threat of a stronger over a weaker species
The bioThreat function below takes two vectors of favourability values at different localities for, respectively, a stronger and a weaker species (e.g., a superior vs. an inferior competitor, or an invasive predator vs. an unadapted native prey), and calculates the level of threat that the former may potentially pose to the latter in each localit...
4614 sym
Plot shared favourability for two competing species
The sharedFav function below implements the graphical analyses of Acevedo et al. (2010, 2012) on biogeographical interactions. It takes two vectors of favourability values at different localities for, respectively, a stronger and a weaker competing species (or two equally strong competitors), and plots their favourableness or shared favourability...
4240 sym 2 img
Variable selection with multGLM
The ‘multGLM‘ function in the ‘fuzzySim‘ R package automatically calculates generalized linear models for one or more species, with a range of options for variable selection. However, you may want to build your final models with other methods (e.g. multimodel-inference with packages ‘glmulti’ or ‘MuMIn’, or other modelling algorit...
1418 sym R (615 sym/1 pcs)
Plot the density of predicted values for presences and absences
The ‘predDensity’ function, included in the modEvA package since version 1.5 (currently available on R-Forge), produces a histogram and/or a kernel density plot of predicted values for observed presences and absences in a binomial GLM: predDensity <- function (model = NULL, obs = NULL, pred = NULL, separate = TRUE, type = c("both"), legend...
644 sym R (3352 sym/2 pcs) 2 img
Area under the precision-recall curve
The AUC function, in the modEvA package, initially computed only the area under the receiver operating characteristic (ROC) curve. Now, since modEvA version 1.7 (currently available on R-Forge), it also offers the option to compute the precision-recall curve, which may be better for comparing models based on imbalanced data (e.g. for rare species...
992 sym R (158 sym/1 pcs) 2 img
modEvA 2.0 now on CRAN!
The new version of modEvA (2.0) is now on CRAN! It can produce some new plots, such as the precision-recall curve and the histograms/densities of predicted values for presence and absence observations. It calculates some additional evaluation measures, such as the area under the precision-recall curve (function ‘AUC‘), mean precision (also fu...
1133 sym R (52 sym/1 pcs)
Grid point occurrence records onto a raster
The ‘gridRecords‘ function, which has just been added to the ‘fuzzySim‘ package (from version 2.6 on), takes a raster stack and a set of spatial coordinates of a species’ presence (and optionally absence) records, and returns a data frame with the presences and absences, as well as the corresponding values of the rasters in the grid of ...
1477 sym R (2705 sym/3 pcs) 4 img
This blog is now on R-bloggers!
I should have done this long ago, but it’s never too late: modTools has finally been added to R-bloggers — a content aggregator contributed by bloggers who write about R, empowering bloggers to empower other R users. I hope the community will find it useful. Constructive feedback welcome! Related To leave a comment for the author, please fo...
680 sym