Bump Hunting for Risk: a New Data Mining Tool and its Applications

Ursula Becker and Ludwig Fahrmeir
Additional contact information
Ursula Becker: University of Munich
Ludwig Fahrmeir: University of Munich

Computational Statistics, 2001, vol. 16, issue 3, No 5, 373-386

Abstract: Summary Bump Hunting is a new data mining technique (Friedman, J. H. & Fisher, N. I. 1999). In this paper we explore its potential for risk assessment. The method is first presented and illustrated by application to credit risk data from a German bank. Based on comparisons with standard analyses of this data set, we conclude that Bump Hunting has potential for identification of risk in financial applications. In the next step the original Bump Hunting algorithm is modified for analysis of censored survival data. This Survival Bump Hunting is used for the analysis of a bone marrow transplant data set and these results are compared to previous analyses which used standard survival methods such as Cox regression. The findings obtained using Survival Bump Hunting confirmed the previous analyses and added some interesting new aspects.

Keywords: Data Mining; Bump Hunting; High-dimensional data; Classification; Credit scoring; Survival analysis; Prognostic factors (search for similar items in EconPapers)
Date: 2001
References: View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s001800100073 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:compst:v:16:y:2001:i:3:d:10.1007_s001800100073

Ordering information: This journal article can be ordered from
http://www.springer.com/statistics/journal/180/PS2

DOI: 10.1007/s001800100073

Access Statistics for this article

Computational Statistics is currently edited by Wataru Sakamoto, Ricardo Cao and Jürgen Symanzik

More articles in Computational Statistics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().