Learn R Programming
⚠️There's a newer version (3.33) of this package.Take me there.

modEvA (version 3.18.2)

Model Evaluation and Analysis

Description

Analyses species distribution models and evaluates their performance. It includes functions for variation partitioning, extracting variable importance, computing several metrics of model discrimination and calibration performance, optimizing prediction thresholds based on a number of criteria, performing multivariate environmental similarity surface (MESS) analysis, and displaying various analytical plots. Initially described in Barbosa et al. (2013) .

Copy Link

Version

Install

install.packages('modEvA')

Monthly Downloads

1,808

Version

3.18.2

License

GPL-3

Maintainer

A. Barbosa

Last Published

August 28th, 2024

Functions in modEvA (3.18.2)

applyThreshold

Apply threshold(s) to model predictions
MESS

Multivariate Environmental Similarity Surfaces based on a data frame
HLfit

Hosmer-Lemeshow goodness of fit
AUC

Area Under the Curve
OA

Overlap Analysis
RsqGLM

R-squared measures for GLMs
Dsquared

Explained deviance
RMSE

Root mean square error
MillerCalib

Miller's calibration satistics for logistic regression models
Boyce

Boyce Index
confusionLabel

Label predictions according to their confusion matrix category
arrangePlots

Arrange plots
getThreshold

Prediction threshold for a given criterion
confusionMatrix

Confusion matrix
inputMunch

Munch inputs into 'obs' and 'pred' vectors
logLike

Log-likelihood
evaluate

Evaluate a model based on the elements of a confusion matrix.
evenness

Evenness in a binary vector.
getModEqn

Get model equation
getBins

Get bins of continuous values.
modEvA-package

Model Evaluation and Analysis
optiThresh

Optimize threshold for model evaluation.
plotGLM

Plot a generalized linear model
predDensity

Plot the density of predicted or predictor values for presences and absences.
optiPair

Optimize the classification threshold for a pair of related model evaluation measures.
multModEv

Multiple model evaluation
predPlot

Plot predicted values for presences and absences, optionally classified according to a prediction threshold.
mod2obspred

Extract observed and predicted values from a model object.
lollipop

Lollipop chart
modEvAmethods

Methods implemented in modEvA functions
quantReclass

Reclassify continuous values based on quantiles
varPart

Variation partitioning
varImp

Variable importance.
range01

Shrink or stretch a vector to make it range between 0 and 1
ptsrast2obspred

Observed and predicted values from presence points and a raster map.
prevalence

Prevalence
similarity

Similarity measures
rotif.mods

Rotifer distribution models
standard01

Standardize to 0-1 (or vice-versa)
threshMeasures

Threshold-based measures of model evaluation