ML Anthology

Pointwise Exact Bootstrap Distributions of ROC Curves

Charles Dugas, David Gadoury
MLJ 2010 pp. 103-136
doi:10.1007/S10994-009-5134-6 /mlj/2010/dugas2010mlj-pointwise/

Abstract

We derive pointwise exact bootstrap distributions of ROC curves and the difference between ROC curves for threshold and vertical averaging. From these distributions, pointwise confidence intervals are derived and their performance is measured in terms of coverage accuracy. Improvements over techniques currently in use are obtained, in particular in the extremes of ROC curves where we show that typical drastic falls in coverage accuracy can be avoided.

PDF MLJ Semantic Scholar

Cite

Text

Dugas and Gadoury. "Pointwise Exact Bootstrap Distributions of ROC Curves." Machine Learning, 2010. doi:10.1007/S10994-009-5134-6

Markdown

[Dugas and Gadoury. "Pointwise Exact Bootstrap Distributions of ROC Curves." Machine Learning, 2010.](https://mlanthology.org/mlj/2010/dugas2010mlj-pointwise/) doi:10.1007/S10994-009-5134-6

BibTeX

@article{dugas2010mlj-pointwise,
  title     = {{Pointwise Exact Bootstrap Distributions of ROC Curves}},
  author    = {Dugas, Charles and Gadoury, David},
  journal   = {Machine Learning},
  year      = {2010},
  pages     = {103-136},
  doi       = {10.1007/S10994-009-5134-6},
  volume    = {78},
  url       = {https://mlanthology.org/mlj/2010/dugas2010mlj-pointwise/}
}