Proceedings chapter
OA Policy
English

Computing and using the deviance with classification trees

Contributors

Ritschard, Gilbert

Presented at

Berlin, 2006

Published in

Rizzi, Alfredo and Vichi, Maurizio (Ed.), COMPSTAT 2006 - Proceedings in Computational Statistics, p. 55-66

Publication date

2006

Abstract

The reliability of induced classification trees is most often evaluated by means of the error rate. Whether computed on test data or through cross-validation, this error rate is suited for classification purposes. We claim that it is, however, a partial indicator only of the quality of the knowledge provided by trees and that there is a need for additional indicators. For example, the error rate is not representative of the quality of the description provided. In this paper we focus on this descriptive aspect. We consider the deviance as a goodness-of-fit statistic that attempts to measure how well the tree is at reproducing the conditional distribution of the response variable for each possible profile (rather than the individual response value for each case) and we discuss various statistical tests that can be derived from them. Special attention is devoted to computational aspects.

Keywords

  • Classification tree
  • Deviance
  • Goodness-of-fit
  • Chi-square statistics
  • BIC

Citation (ISO format)

RITSCHARD, Gilbert. Computing and using the deviance with classification trees. In: COMPSTAT 2006 - Proceedings in Computational Statistics. Rizzi, Alfredo and Vichi, Maurizio (Ed.). Berlin. [s.l.] : [s.n.], 2006. p. 55–66. doi: 10.1007/978-3-7908-1709-6_5

Main files (1)

Proceedings chapter

Identifiers

665views
945downloads

Technical informations

Creation

01/12/2009 15:28:11

First validation

01/12/2009 15:28:11

Update

14/03/2023 15:19:10

Status update

14/03/2023 15:19:09

Last indexation

29/10/2024 12:45:06
All rights reserved by Archive ouverte UNIGE and the University of Geneva
OpenAIRE Indexed & Validated