ON THE SELECTED METHODS FOR EVALUATING CLASSIFICATION MODELS

Authors

  • Małgorzata Misztal

Keywords:

classifier performance, predictive ability, ROC curve, reclassification, decision curve

Abstract

Traditional measures for assessing the performance of classification models for binary outcomes are the ROC curve and the area under the ROC curve (AUC).

Reclassification tables (Cook, 2008), net reclassification improvement (NRI) and integrated discrimination improvement (IDI) (Pencina et al., 2008) or decision – analytic measures with decision curve analysis (Vickers & Elkin, 2006) have been recently proposed for evaluating the predictive ability of classifiers.

This paper analyzes the measures mentioned above with some credit taking application.

 

Downloads

Download data is not yet available.

Downloads

Published

2014-11-09

How to Cite

Misztal, M. (2014). ON THE SELECTED METHODS FOR EVALUATING CLASSIFICATION MODELS. Acta Universitatis Lodziensis. Folia Oeconomica, 3(302). Retrieved from https://czasopisma.uni.lodz.pl/foe/article/view/59

Issue

Section

Economics

Similar Articles

1 2 3 4 5 6 > >> 

You may also start an advanced similarity search for this article.