CART analysis of qualitative variables to improve credit rating processes

Giampaolo Gabbi (), Massimo Matthias and Marco De Lerma
Additional contact information
Massimo Matthias: University of Siena, Italy
Marco De Lerma: University of Siena, Italy

No 179, Computing in Economics and Finance 2006 from Society for Computational Economics

Abstract: Qualitative behaviours of small firms are explored to forecast criticalities in the Italian credit market. We build up an evaluation process to integrate quantitative rating practices. Research method: Relevant qualitative factors to estimate the credit risk are empirically investigated by choosing variables related to seven perspectives of analysis (sector, governance, internal processes, learning and growth, customers, economic-financial analysis, quality of balance sheet). Outcomes have been elaborated through the Classification And Regression Tree. Rating evaluation is based upon qualitative factors. Findings: The capability to forecast bad firms is 92.4\%, while 84.5\% is the percentage to forecast good firms. The Cumulative Accuracy Curve shows the 84\% ability to explain the variance of phenomenon. Main results: ratings of firms are primary explained by aptitude to reach short and long term purposes; banks’ analysts should integrate their quantitative models with qualitative data; methodologies employed offer a quantitative solution to estimate the weight of each variable. Conclusions: When balance sheets are characterized by small consistency, qualitative variables should be taken into consideration to elaborate or integrate rating procedures.

Keywords: credit risk; qualitative models; rating; regression tree (search for similar items in EconPapers)
JEL-codes: G21 C25 (search for similar items in EconPapers)
Date: 2006-07-04

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

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: http://EconPapers.repec.org/RePEc:sce:scecfa:179

Access Statistics for this paper

More papers in Computing in Economics and Finance 2006 from Society for Computational Economics
Contact information at EDIRC.
Series data maintained by Christopher F. Baum ().