predDensity {modEvA}

Plot the density of predicted values for presences and absences.

Description

This function produces a histogram and/or a kernel density plot of predicted values for a binary-response model, possibly separately for the observed presences and absences, given a model object or a vector of predicted values and (optionally) a vector of the corresponding observed values. When there are multiple predicted values for each site, it can also plot a confidence interval.

Usage

predDensity(model = NULL, obs = NULL, pred = NULL,
separate = TRUE, type = "both", ci = NA, legend.pos = "topright",
main = "Density of predicted values", na.rm = TRUE, rm.dup = FALSE,
xlim = NULL, ...)

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 'pred' (and optionally 'obs') argument(s) instead of 'model'.

obs

alternatively to 'model' and together with 'pred', an optional numeric vector (in the same order of 'pred') 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 may be omitted (to show the density plot of all 'pred' values combined), and it is ignored if 'model' is provided.

pred

alternatively to 'model', a vector of predicted values of presence probability, habitat suitability, environmental favourability or alike. Must be of the same length and in the same order as 'obs' (if the latter is provided). 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.

separate

logical value indicating whether prediction densities should be computed separately for observed presences (ones) and absences (zeros). Defaults to TRUE, but it is automatically changed to FALSE if either 'model' or 'obs' are not provided, or if 'ci' is not NULL.

type

character vector specifying whether to produce a "histogram", a "density" plot, or "both" (the default). Partial argument matching is used.

ci

numeric value for a confidence interval to add to the plot, e.g. 0.95 for 95%. The default is NA.

legend.pos

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

main

main title for the plot.

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.

xlim

numeric vector of length 2 setting the limits for the x axis of the plot. The default is NULL, for the range of the density of predicted values.

...

additional arguments to pass to hist, e.g. 'breaks' or 'border'.

Details

For more details, please refer to the documentation of the functions mentioned under "See Also".

Value

This function outputs and plots the object(s) specified in 'type' – by default, a density object and a histogram.

Author(s)

A. Marcia Barbosa

See Also

hist, density, predPlot

Examples

# load sample models:
data(rotif.mods)

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


# compute predDensity with different parameters:

predDensity(model = mod)

predDensity(model = mod, breaks = seq(0, 1, by = 0.05))

predDensity(model = mod, type = "histogram")

predDensity(model = mod, type = "density")

predDensity(model = mod, ci = 0.975)


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

obs <- mod$y
pred <- mod$fitted.values

predDensity(obs = obs, pred = pred)

predDensity(pred = pred, ci = 0.95)


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