movieROC - Visualization for Classification Regions for ROC Curve
Generalization
Tools for estimating the Receiver Operating Characteristic
(ROC) curve under different generalizations: - making the
classification subsets flexible to cover those scenarios where
both extremes of the marker are associated with a higher risk
of being positive, considering two thresholds (gROC curve); -
transforming the marker by a function either defined by the
user or resulting from a logistic regression model (hROC
curve); - considering a linear transformation with some fixed
parameters introduced by the user, dynamic parameters resulting
from Meisner et al. (2017) approach or empirically maximizing
TPR for each FPR for a bivariate marker. Also a quadratic
transformation with particular coefficients or a function
fitted by a logistic regression model can be considered (biROC
curve); - considering a linear transformation with some fixed
parameters introduced by the user, dynamic parameters resulting
from Meisner et al. (2017) approach or a function fitted by a
logistic regression model (multiROC curve). The classification
regions behind each point of the ROC curve are displayed in
both fixed graphics (plot.buildROC, plot.regions or
plot.funregions function) or videos (movieROC function).