Model selection with Gini indices under auto-calibration
Machine Learning
2022-08-11 v3 Machine Learning
Abstract
The Gini index does not give a strictly consistent scoring rule in general. Therefore, maximizing the Gini index may lead to wrong decisions. The main issue is that the Gini index is a rank-based score that is not calibration-sensitive. We show that the Gini index allows for strictly consistent scoring if we restrict to the class of auto-calibrated regression models.
Cite
@article{arxiv.2207.14372,
title = {Model selection with Gini indices under auto-calibration},
author = {Mario V. Wüthrich},
journal= {arXiv preprint arXiv:2207.14372},
year = {2022}
}
Comments
11 pages, 1 figure