ROC Curves

Receiver Operating Characteristic curves

See also:

• Logistic Regression

• Sensitivity and Specificity

• AUC (Area Under the Curve)

• Type 1 error and Type 2 error

• Wikipedia ROC curve entry

• Attachments below (presentation, document with Wikipedia info)

R code for demonstration

library(TeachingDemos)

if(interactive()){

roc.demo()

with(CO2,

roc.demo(uptake[Type=='Mississippi'],

uptake[Type=='Quebec'] )

)

}

R packages

Several R packages allow the creation and use of ROC curves. Only a few are listed below.

epicalc package ('lroc' function)

rocr package

pROC package (also here)

This package appears to allow comparison of ROC curves and calculation of confidence intervals for estimates

* pROC (1.2.1)

Xavier Robin

http://crantastic.org/packages/pROC

Tools for visualizing, smoothing and comparing receiver operating

characteristic (ROC curves). (Partial) area under the curve (AUC)

can be compared with statistical tests based on U-statistics or

bootstrap. Confidence intervals can be computed for (p)AUC or ROC

curves.