Download this Paper Open PDF in Browser

The Recursive Partitioning Algorithm (RPA): A Nonparametric Classification System

10 Pages Posted: 16 Sep 2012  

Luiz C. J. Perera

Mackenzie Presbyterian University

Roberto Borges Kerr

Mackenzie Presbyterian University

Herbert Kimura

Mackenzie Presbiterian University

Fabiano Guasti Lima

University of Sao Paulo (USP)

Date Written: September 14, 2012

Abstract

Leo Breiman (Breiman et al., 1984, 1998) was a statistician who was fond of practical applications, and this led him to develop several original studies. Based on the work begun by Friedman (1977), he developed a very accurate classification system, without the need for statistical assumptions, since it is a nonparametric methodology. The aim of this study is to present the work of Breiman known as the Recursive Partitioning Algorithm. The RPA will be introduced as a nonparametric approach to credit analysis, allowing for the incorporation of the costs of misclassifications. Several studies, such as Novak and LaDue (1999) and Marais, Patais and Wolfson (1984), have shown its applicability in the analysis and granting of credit. A long road has been traveled from the early work of Friedman (1977) to the CART model developed by Steinberg and Golovnya (2006). This paper – apart from presenting the fundamentals and possibilities for use of the RPA – seeks to show the effectiveness of the results attained through a comparison with a parametric model, the Discriminant Analysis, considered the most traditional and classical method of analysis. The results show the RPA to be a superior technique, as well as a technique of easy intuition by analysts. The conclusion of the paper confirms that the RPA system – little known and discussed by academics and market professionals – is a powerful classificatory tool, with the advantage of being nonparametric.

Keywords: Recursive Partitioning Algorithm (RPA), Classification and Regression Trees (CART), Classificatory Models, Nonparametric Methodology, Discriminant Analysis

JEL Classification: C38, G17, G21, G24, G32, G33

Suggested Citation

Perera, Luiz C. J. and Kerr, Roberto Borges and Kimura, Herbert and Lima, Fabiano Guasti, The Recursive Partitioning Algorithm (RPA): A Nonparametric Classification System (September 14, 2012). Available at SSRN: https://ssrn.com/abstract=2146964 or http://dx.doi.org/10.2139/ssrn.2146964

Luiz Carlos Jacob Perera (Contact Author)

Mackenzie Presbyterian University ( email )

Rua da Consolação 896
São Paulo, São Paulo 010302907
Brazil
+55.11.21148273 (Phone)
+5511.21148836 (Fax)

HOME PAGE: http://www.jperera.pro.br

Roberto Borges Kerr

Mackenzie Presbyterian University ( email )

Rua da Consolação 920
São Paulo, 010302907
Brazil

Herbert Kimura

Mackenzie Presbiterian University ( email )

Rua da Consolacao, 896 - 6o andar
01302-907 Sao Paulo
Brazil
+55 11 3236-8589 (Phone)
+55 11 3236-8600 (Fax)

Fabiano Guasti Lima

University of Sao Paulo (USP) ( email )

Rua Luciano Gualberto, 315
São Paulo, São Paulo 14800-901
Brazil

Paper statistics

Downloads
69
Rank
284,399
Abstract Views
553