Recursive Partitioning

S. Landau, S. Barthel

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

1 Citation (Scopus)

Abstract

This article provides an introduction to recursive partitioning techniques; that is, methods that predict the value of a response variable by forming subgroups of subjects within which the response is relatively homogeneous on the basis of the values of a set of predictor variables. Resulting rules are conveniently summarized by a decision tree referred to as a classification tree for categorical responses and as a regression tree for continuous responses. We demonstrate the steps involved in the construction of such trees by using two data sets, one on Irish school children and another on leading companies.

Original languageEnglish
Title of host publicationInternational Encyclopedia of Education, Third Edition
PublisherElsevier
Pages383-389
Number of pages7
ISBN (Electronic)9780080448947
DOIs
Publication statusPublished - 1 Jan 2009

Fingerprint

Dive into the research topics of 'Recursive Partitioning'. Together they form a unique fingerprint.

Cite this