Corporate credit default models: a mixed logit approach

Martin Kukuk () and Michael Rönnberg ()

Review of Quantitative Finance and Accounting, 2013, vol. 40, issue 3, 467-483

Abstract: The popular logit model is extended to allow for varying stochastic parameters (mixed logit) and non-linearities of regressor variables while analysing a cross-sectional sample of German corporate credit defaults. With respect to economic interpretability and goodness of probability forecasts according to disriminatory power and calibration, empirical results favor the extended specifications. The mixed logit model is particularly useful with respect to interpretability. However, probability forecasts based on the mixed logit model are not distinctively preferred to extended logit models allowing for non-linearities in variables. Further potential improvements with the help of the mixed logit approach for panel data are shown in a Monte Carlo study. Copyright Springer Science+Business Media, LLC 2013

Keywords: Credit default models; Binary response models; Model specification; Estimation of probabilities of default; Mixed logit; C52; G24 (search for similar items in EconPapers)
Date: 2013
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (5)

Downloads: (external link)
http://hdl.handle.net/10.1007/s11156-012-0281-4 (text/html)
Access to full text is restricted to subscribers.

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:kap:rqfnac:v:40:y:2013:i:3:p:467-483

Ordering information: This journal article can be ordered from
http://www.springer.com/finance/journal/11156/PS2

DOI: 10.1007/s11156-012-0281-4

Access Statistics for this article

Review of Quantitative Finance and Accounting is currently edited by Cheng-Few Lee

More articles in Review of Quantitative Finance and Accounting from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().