On Bootstrapping the ROC Curve

Abstract

This paper is devoted to thoroughly investigating how to bootstrap the ROC curve, a widely used visual tool for evaluating the accuracy of test/scoring statistics in the bipartite setup. The issue of confidence bands for the ROC curve is considered and a resampling procedure based on a smooth version of the empirical distribution called the smoothed bootstrap" is introduced. Theoretical arguments and simulation results are presented to show that the "smoothed bootstrap" is preferable to a "naive" bootstrap in order to construct accurate confidence bands."

Cite

Text

Bertail et al. "On Bootstrapping the ROC Curve." Neural Information Processing Systems, 2008.

Markdown

[Bertail et al. "On Bootstrapping the ROC Curve." Neural Information Processing Systems, 2008.](https://mlanthology.org/neurips/2008/bertail2008neurips-bootstrapping/)

BibTeX

@inproceedings{bertail2008neurips-bootstrapping,
  title     = {{On Bootstrapping the ROC Curve}},
  author    = {Bertail, Patrice and Clémençcon, Stéphan J. and Vayatis, Nicolas},
  booktitle = {Neural Information Processing Systems},
  year      = {2008},
  pages     = {137-144},
  url       = {https://mlanthology.org/neurips/2008/bertail2008neurips-bootstrapping/}
}