ROC-curves are a very common tool to compare tests (in medical/clinical applications) or prediction models in general. Some time ago a former colleague Prof. Dr. Dr. meintrup.davidwrote a nice script that does ROC- and pAUC-analysis in JMP.
This is the latest version of the script and it is giving you:
- ROC-curves for one or more tests
- Bootstrap-, Binomial- and Kolmogroff-Confidence Bands for the ROC-curves
- Weighted cutpoints
- parcial Area under the Curve (pAUC)
I have started to write a mini-series on our Blog about this add-in and how you can use it: Statcon Comparing ROC-curves in JMP