roc_plot_format: Format data prior to plotting ROC plot for one or more...

In kamclean/predictr: Facilitates development and validation of predictive models

Format data prior to plotting ROC plot for one or more models.

Description

Format data prior to plotting ROC plots for one or more models.

Usage

roc_plot_format(
  data,
  model = NULL,
  event = "event",
  predict = "predict",
  smooth = F,
  confint = F
)

Arguments

data	Dataframe with at least 2 columns: (1) binary outcome of interest, (2) predicted probability of outcome (binary or continuous)
model	Optional string specifying the column name specifying which model each outcome/prediction belongs to (default ="model"). This allows multiple models to be assessed within 1 function.
event	String specifying the column name containing the binary outcome of interest (default ="event")
predict	String specifying the column name containing predicted probability of outcome (default ="prediction")
smooth	Logical value specifying if the ROC curve should be smoothed (default=FALSE)
confint	Logical value specifying if the confidence interval of the sensitivity/specificity should be calculated (default = FALSE)

Value

Tibble with sensitivity/specificity for one or more models.

kamclean/predictr documentation built on Aug. 14, 2022, 4:35 a.m.