Could the Trigger to the Subprime Crisis Have Been Predicted? A Mortgage Risk Modeling Approach
Jose Molina Utrilla
University of Essex - Centre for Computational Finance and Economic Agents
University of Essex - Essex Business School
November 22, 2010
The abnormally high mortgage default rates that became apparent in early 2007 were not foreseen in June 2005, when mortgage production in the US reached its peak. Could the significant increase in mortgage defaults that triggered the resultant subprime crisis, have been predicted? This paper develops a mortgage-level predictive model for mortgage default and delinquency rates, based on a logistic regression and Markov chain framework. The results are compared against actual fixed rate mortgage-level default data and provide strong evidence that the high US nonprime mortgage default rates were predictable in mid-2005 using historical data only available at the time.
Number of Pages in PDF File: 31
Keywords: Default Probabilities, Logistic Regression, Markov Chain, Credit Risk
JEL Classification: G01, G17, G32working papers series
Date posted: May 28, 2010 ; Last revised: April 24, 2011
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo3 in 0.328 seconds