This file is part of IDEAS, which uses RePEc data


[ Papers | Articles | Software | Books | Chapters | Authors | Institutions | JEL Classification | NEP reports | Search | New papers by email | Author registration | Rankings | Volunteers | FAQ | Blog | Help! ]

Variable Selection in Data Mining: Building a Predictive Model for Bankruptcy

Author info | Abstract | Publisher info | Download info | Related research | Statistics
Author Info
Dean P. Foster
Robert A. Stine

Additional information is available for the following registered author(s):

Abstract

We develop and illustrate a methodology for fitting models to large, complex data sets. The methodology uses standard regression techniques that make few assumptions about the structure of the data. We accomplish this with three small modifications to stepwise regression: (1) We add interactions to capture non-linearities and indicator functions to capture missing values; (2) We exploit modern decision theoretic variable selection criteria; and (3) We estimate standard error using a conservative approach that works for heteroscedastic data. Omitting any one of these modifications leads to poor performance.

We illustrate our methodology by predicting the onset of personal bankruptcy among users of credit cards. This applications presents many challenges, ranging from the rare frequency of bankruptcy to the size of the available database. Only 2,244 bankruptcy events appear among some 3 million months of customer activity. To predict these, we begin with 255 features to which we add missing value indicators and pairwise interactions that expand to a set of over 67,000 potential predictors. From these, our method selects a model with 39 predictors chosen by sequentially comparing estimates of their significance to a series of thresholds. The resulting model not only avoids over-fitting the data, it also predicts well out of sample. To find half of the 1800 bankruptcies hidden in a validation sample of 2.3 million observations, one need only search the 8500 cases having the largest model predictions.

Download Info
To download:

If you experience problems downloading a file, check if you have the proper application to view it first. Information about this may be contained in the File-Format links below. In case of further problems read the IDEAS help page. Note that these files are not on the IDEAS site. Please be patient as the files may be large.

File URL: http://fic.wharton.upenn.edu/fic/papers/01/0105.pdf
File Format: application/pdf
File Function:
Download Restriction: no

Publisher Info
Paper provided by Wharton School Center for Financial Institutions, University of Pennsylvania in its series Center for Financial Institutions Working Papers with number 01-05.

Download reference. The following formats are available: HTML (with abstract), plain text (with abstract), BibTeX, RIS (EndNote, RefMan, ProCite), ReDIF
Length:
Date of creation: Feb 2001
Date of revision:
Handle: RePEc:wop:pennin:01-05

Contact details of provider:
Postal: 3301 Steinberg Hall-Dietrich Hall, 3620 Locust Walk, Philadelphia, PA 19104.6367
Phone: 215.898.1279
Fax: 215.573.8757
Email:
Web page: http://fic.wharton.upenn.edu/fic/
More information through EDIRC

For technical questions regarding this item, or to correct its listing, contact: (Thomas Krichel).

Related research
Keywords:

This paper has been announced in the following NEP Reports:

Statistics
Access and download statistics

Did you know? About 2700 working paper series are listed on RePEc.

This page was last updated on 2009-12-6.

This information is provided to you by IDEAS at the Department of Economics, College of Liberal Arts and Sciences, University of Connecticut using RePEc data on a server sponsored by the Society for Economic Dynamics.