modEvA (version 3.5)

predPlot: Plot predicted values for presences and absences, classified according to a prediction threshold.

Description

This function plots predicted values separated into observed presences and absences and coloured according to whether they are above or below a given prediction threshold. The plot imitates (with permission from the author) one of the graphical outputs of the 'summary' of models built with the embarcadero package (Carlson, 2020), but it can be applied to any 'glm' object or any set of observed and predicted values, and it allows specifying a user-defined threshold.

Usage

predPlot(model = NULL, obs = NULL, pred = NULL, thresh = "preval", 
main = "Classified predicted values", legend.pos = "n", pch = 1, cex = 0.5, 
col = c("black", "grey"), na.rm = TRUE, rm.dup = FALSE, interval = 0.01, 
quant = 0)

Value

This function outputs a plot as per 'Description'.

Arguments

model

a binary-response model object of class "glm", "gam", "gbm", "randomForest" or "bart". If this argument is provided, 'obs' and 'pred' will be extracted with mod2obspred. Alternatively, you can input the 'obs' and 'pred' arguments instead of 'model'.

obs

alternatively to 'model' and together with 'pred', a numeric vector of observed presences (1) and absences (0) of a binary response variable. Alternatively (and if 'pred' is a 'SpatRaster'), a two-column matrix or data frame containing, respectively, the x (longitude) and y (latitude) coordinates of the presence points, in which case the 'obs' vector will be extracted with ptsrast2obspred. This argument is ignored if 'model' is provided.

pred

alternatively to 'model' and together with 'obs', a vector with the corresponding predicted values of presence probability, habitat suitability, environmental favourability or alike. Must be of the same length and in the same order as 'obs'. Alternatively (and if 'obs' is a set of point coordinates), a 'SpatRaster' map of the predicted values for the entire evaluation region, in which case the 'pred' vector will be extracted with ptsrast2obspred. This argument is ignored if 'model' is provided.

thresh

threshold value to separate predicted presences from predicted absences in 'pred'; can be "preval" (the default), to use the prevalence (i.e. proportion of presences) in 'obs', or any real number between 0 and 1. This value will be used to draw a vertical line on the plot and to colour the points (predicted values) according to whether they fall below or above the threshold. Run modEvAmethods("getThreshold") for available options, and see Details in getThreshold for their description.

main

Main title for the plot.

legend.pos

character value specifying the position for the legend on the plot. Can be "bottomleft", "bottom", "bottomright", "topleft", "left", "top", "topright", "right", "center", or NA or "n" for no legend (the default). Partial argument matching is used.

pch

plotting character for the presences and absences (see par).

cex

relative size of the plotting character (see par).

col

vector of length 2 indicating the colours with which to plot predicted presences and absences (points above and below the threshold), respectively.

na.rm

Logical value indicating whether missing values should be ignored in computations. Defaults to TRUE.

rm.dup

If TRUE and if 'pred' is a SpatRaster and if there are repeated points within the same pixel, a maximum of one point per pixel is used to compute the presences. See examples in ptsrast2obspred. The default is FALSE.

interval

Argument to pass to optiThresh indicating the interval between the thresholds to test, if 'thresh' implies optimizing a threshold-based measure. The default is 0.01. Smaller values may provide more precise results but take longer to compute.

quant

Numeric value indicating the proportion of presences to discard if thresh="MTP" (minimum training presence). With the default value 0, MTP will be the threshold at which all observed presences are classified as such; with e.g. quant=0.05, MTP will be the threshold at which 5% presences will be classified as absences.

Author

A. Marcia Barbosa

References

Carlson C.J. (2020) embarcadero: Species distribution modelling with Bayesian additive regression trees in R. Methods in Ecology and Evolution, 11: 850-858.

See Also

predDensity, plotGLM

Examples

# load sample models:
data(rotif.mods)

# choose a particular model to play with:
mod <- rotif.mods$models[[1]]

predPlot(model = mod)

predPlot(model = mod, thresh = 0.5)


# you can first select a threshold optimized according to a particular metric:

if (FALSE) {
threshold <- optiThresh(mod, measures = "TSS", optimize = "each")
threshold <- threshold$optimals.each[ , "threshold"]
threshold
predPlot(model = mod, thresh = threshold)
}

# you can also use 'predPlot' with vectors of observed and predicted values
# instead of a model object:

presabs <- mod$y
prediction <- mod$fitted.values

predPlot(obs = presabs, pred = prediction)

predPlot(obs = presabs, pred = prediction, thresh = 0.5)


# 'obs' can also be a table of presence point coordinates
# and 'pred' a SpatRaster of predicted values