dropROC: ROC Curves for all Single Term Deletions from a Model
Description
Given a fitted point process model,
consider dropping each possible term in the model,
and compute the ROC curve for the dropped covariate.
Usage
dropROC(object, scope = NULL, high=TRUE, ...)
Value
A named list containing the ROC curves for each possible deletion.
The individual entries belong to class "fv",
so they can be plotted.
The list belongs to the class "anylist"
so it can be plotted in its entirety.
Arguments
object
A fitted point process model (object of class "ppm",
"kppm", "dppm", "slrm" or "lppm").
scope
A formula or a character vector specifying the terms to be
considered for deletion. The default is all possible terms.
high
Argument passed to roc to specify whether the
ROC curves should be based on high or low values of the covariates.
Either a logical value, or a logical vector of the same length
as scope.
...
Arguments passed to as.mask to control the
pixel resolution used for calculation.
This function is like drop1
in that it considers each possible term in the model object
(or only the terms listed in the scope argument),
deletes each such term from the model, and measures the change in the model.
In this case the change is measured by computing the ROC curve
for the deleted covariate, using the updated model as a baseline.
References
Baddeley, A., Rubak, E., Rakshit, S. and Nair, G. (2025) ROC curves for spatial point patterns and presence-absence data. tools:::Rd_expr_doi("10.48550/arXiv.2506.03414").
dimyx <- if(interactive()) NULLelse32 fut <- ppm(bei ~ grad + elev, data=bei.extra)
z <- dropROC(fut, dimyx=dimyx)
plot(z)
## how to compute AUC for each sapply(z, auc)