An Ordered Qualitative Response Modeling Approach for the Estimation of Corporate Defaults and Other Forms of Exit
Abstract
We propose a new approach for the estimation of defaults and other forms of exit of borrowers. Our approach is based on the ordered qualitative response model. We first show that any ordered qualitative response model is equivalent to the competing risks model – commonly employed in the estimation of corporate defaults and other forms of exit – in continuous-time. We then construct the continuous-time likelihood function of the models and further present its discrete-time simplification. Lastly, we compare and contrast the competing risks and ordered qualitative response models through numerical experiments in a two-state setting, and demonstrate that none of the alternatives necessarily dominates the others. Our results indicate that it may be worthwhile to estimate the models in continuous-time.
Keywords
References
- AMEMIYA, Takeshi, “Qualitative Response Models”, Annals of Economic and Social Management, 4 (3), 1975, pp. 363-372.
- DAS, Sanjiv, Darrel DUFFIE, Nikunj KAPADIA, and Leandro SAITA, “Common Failings: How Corporate Defaults are Correlated”, Journal of Finance, 62, 2007, pp. 93–117.
- DUFFIE, Darrel, Andreas ECKNER, Guillaume HOREL, and Leandro SAITA, “Frailty Correlated Default”, Journal of Finance, 64, 2009, pp. 2089–2123.
- DUFFIE, Darrel, Leandro SAITA, and Ke WANG, “Multi-period Corporate Default Prediction with Stochastic Covariates”, Journal of Financial Economics, 83 (3), 2007, pp. 635–665.
- SHUMWAY, Tyler, “Forecasting Bankruptcy More Accurately: A Simple Hazard Model”, Journal of Business, 74, 2001, pp. 101–124.
Details
Primary Language
English
Subjects
-
Journal Section
Research Article
Authors
Sinan Alçın
Istanbul Kultur University, Department of Economics
Türkiye
T. Sabri Öncü
This is me
New York University, Stern School of Business
United States
Publication Date
October 31, 2017
Submission Date
September 27, 2016
Acceptance Date
July 31, 2017
Published in Issue
Year 2017 Volume: 4 Number: 2