ROC curve

Live demo

Open this example in Shinylive

The app runs entirely in the browser with webR; no local R installation is required for the demo.

This demo uses two Normal inputs on the same fixed domain. The common scale makes changes in location and spread directly comparable.

The statistical calculations are base R: dnorm() for the score densities and pnorm() for the ROC curve.

score_domain <- c(-5, 5)

modist_input(
  "positive",
  value = list(mu = 1, sigma = 1),
  domain = score_domain
)

modist_input(
  "negative",
  value = list(mu = -1, sigma = 1),
  domain = score_domain
)

Run the complete sidebar example with:

shiny::runApp(system.file("examples/roc", package = "shinymodist"))

The example uses bslib only for layout. The probability calculations and plots remain base R.