Ensemble predictions of recovery rates

João Bastos

No 1301, CEMAPRE Working Papers from Centre for Applied Mathematics and Economics (CEMAPRE), School of Economics and Management (ISEG), Technical University of Lisbon

Abstract: In many domains, the combined opinion of a committee of experts provides better decisions than the judgment of a single expert. This paper shows how to implement a successful ensemble strategy for predicting recovery rates on defaulted debts. Using data from Moody's Ultimate Recovery Database, it is shown that committees of models derived from the same regression method present better forecasts of recovery rates than a single model. More accurate predictions are observed whether we forecast bond or loan recoveries, and across the entire range of actual recovery values.

Keywords: Recovery rate; Loss given default; Forecasting; Ensemble learning; Credit risk (search for similar items in EconPapers)
JEL-codes: G17 G21 (search for similar items in EconPapers)
Pages: 26 pages
Date: 2013-03
New Economics Papers: this item is included in nep-ban and nep-for
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
http://cemapre.iseg.utl.pt/RePEc/papers/WP1301.pdf (application/pdf)
Our link check indicates that this URL is bad, the error code is: 500 Can't connect to cemapre.iseg.utl.pt:80 (No such host is known. )

Related works:
Journal Article: Ensemble Predictions of Recovery Rates (2014)
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:cma:wpaper:1301

Access Statistics for this paper

More papers in CEMAPRE Working Papers from Centre for Applied Mathematics and Economics (CEMAPRE), School of Economics and Management (ISEG), Technical University of Lisbon Contact information at EDIRC.
Bibliographic data for series maintained by Helena Lima ( this e-mail address is bad, please contact ).