文档介绍：SAS APAUGC 2006 MUMBAI
Comparing decision trees with logistic regression for credit
risk analysis
S S Satchidananda Jay
Research Director & Professor, CBIT, ABIBA Systems
International Institute of Information Bangalore, India
Technology, Bangalore, India @
sssatchidananda@
Abstract
Credit risk evaluation is an important and interesting problem in financial
analysis domain. Several techniques like expert systems, works etc.
have been used for credit rating. However these methods have limitations of
knowledge bottleneck, slow learning etc. Recently decision trees have been
proposed as the white-box models for learning and classification. In this work
an attempt has been made to evaluate decision tree learning scheme with a
logistic regression classifier on default risk of agricultural loans. It has been
found that the decision tree classifiers will produce good results with
parsimonious models.
1. Introduction
Credit-risk evaluation is a very challenging and important data mining problem in the
domain of financial analysis. Many classification methods have been suggested in the
literature to tackle this problem. But most of them are not accepted by the practicing
experts due to various reasons. Against this background, we examine two classifiers in
terms of their accuracy, True Positive and False Negatives w