The Ooghe-Joos-De Vos failure prédiction models: a cross-industry validation

Hubert Ooghe, Sofie Balcaen and Jan Camerlynck

Brussels Economic Review, 2003, vol. 46, issue 1, 39-70

Abstract: This study compares the predictive performances of the Ooghe-Joos-De Vos models across different industries and subgroups concerning size-class and annual accounts form. Type I, type II and unweighted error rates, Gini-coefficients and trade-off functions are analysed. The results indicate a wide variety of performances for the different subgroups. Firstly, both OJD models seem to perform best for the classical manufacturing industries, while they are poorer predictors for the service industries. Secondly, the short-term OJD model has the best predictive abilities for large companies and companies with a complete annual accounts form. In general, the performance differences of the short-term model are much larger than those of the medium-term model and, on average, the medium-term model shows a lower classification accuracy than the short-term model due to a stronger tendency to misclassify non-failing firms as failing.

Keywords: failure prédiction model; validation; performance measures; industry; size (search for similar items in EconPapers)
JEL-codes: G33 M49 (search for similar items in EconPapers)
Date: 2003
References: Add references at CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://dipot.ulb.ac.be/dspace/bitstream/2013/11855/1/ber-0247.pdf ber-0247 (application/pdf)

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:bxr:bxrceb:y:2003:v:46:i:1:p:39-70

Ordering information: This journal article can be ordered from
http://hdl.handle.ne ... ulb.ac.be:2013/11855

Access Statistics for this article

More articles in Brussels Economic Review from ULB -- Universite Libre de Bruxelles Contact information at EDIRC.
Bibliographic data for series maintained by Benoit Pauwels ().