Commercial Mortgage Default: A Comparison of Logit with Radial Basis Function Networks

Posted: 2 Jul 1998

See all articles by Andreas Pericli

Andreas Pericli

ALM & Fixed Income Research

Athanasios Episcopos

Athens University of Economics and Business - Department of Accounting and Finance

Jianxun Hu

Advanta Mortgage

Multiple version iconThere are 2 versions of this paper

Date Written: September 1, 1995

Abstract

This manuscript explores the use of artificial neural networks in the modeling of foreclosure of commercial mortgages. The study employs a large set of individual loan histories previously used in the literature of proportional hazard models on loan default. Radial basis function networks are trained on the same inputs as those used in the logistic, and performance is assessed in terms of prediction accuracy. Neural networks are shown to be superior to the logistic benchmark in terms of discriminating between "good" and "bad" loans. Sensitivity analysis performed on the average loan demonstrates the use of neural networks as an analytical tool. Finally, the study offers suggestions on further improving prediction of defaulting loans.

JEL Classification: G21

Suggested Citation

Pericli, Andreas and Episcopos, Athanasios and Hu, Jianxun, Commercial Mortgage Default: A Comparison of Logit with Radial Basis Function Networks (September 1, 1995). Available at SSRN: https://ssrn.com/abstract=9048

Andreas Pericli

ALM & Fixed Income Research ( email )

8200 Jones Branch Drive
Mailstop 375
Mclean, VA 22102
United States
703-903-2429 (Phone)
Not available (Fax)

Athanasios Episcopos (Contact Author)

Athens University of Economics and Business - Department of Accounting and Finance ( email )

76 Patission Street
GR-104 34 Athens
Greece
+30 21 0820 3364 (Phone)
+30 21 0822 8816 (Fax)

HOME PAGE: http://www.aueb.gr/users/episcopos

Jianxun Hu

Advanta Mortgage ( email )

850 Ridgeview Drive
Horsham, PA 19044
215-576-7646 (Phone)
215-444-4697 (Fax)

Register to save articles to
your library

Register

Paper statistics

Abstract Views
1,015
PlumX Metrics