rocm: Builds the ROC movie (ROCM) an animated sequence of ROC...

In evwalz/uroc: Computes ROCM, UROC and CPA

Builds the ROC movie (ROCM) an animated sequence of ROC curves.

Description

This function computes the sequence of ROC curves which form the ROC Movie and produces a GIF animated ROCM.

Usage

rocm(
  response,
  predictor,
  a = NULL,
  b = NULL,
  object = FALSE,
  gif = TRUE,
  movie.name = "animation.gif",
  ...
)

Arguments

response	a numeric vector of real valued responses.
predictor	a numeric vector of the same length than response, containing real valued predictions for each observation.
a	selects a subset of all ROC curves for the ROC movie with at least a and at most a+b ROC curves.
b	selects a subset of all ROC curves for the ROC movie with at least a and at most a+b ROC curves.
object	if TRUE a list of ROC curves is returned (default object = TRUE).
gif	if TRUE a gif animation is created.
movie.name	name of the movie (with extension).
...	parameters to control the behavior of the GIF animation using the external function ani.option from animation.

Details

The ROC movie can be used to visualize the performance of a real valued foreacsting problem. Therefore, a sequence of ROC curves is generated which can than be combined into a GIF animation. Each entry of the list consist of two vectors of length 1000 containing the values of farate (1-Specificity) and hitrate (sensitivity) and three values, namely the associated auc value, the weight and the threshold.

Value

if object = TRUE, this function returns a list of ROC curves.

Examples

## Not run: 
data(longley)
response = longley$Employed
predictor = longley$GNP
rocm(response, predictor)
## End(Not run)

evwalz/uroc documentation built on May 22, 2022, 1:40 a.m.