Title:	PREDICTING MORTGAGE DEFAULT: LESSONS FROM DATA MINING FANNIE MAE MORTGAGE PORTFOLIO
Author(s):	Stanislav Mamonov, Raquel Benbunan-Fich
ISBN:	978-989-8533-54-8
Editors:	Piet Kommers and Guo Chao Peng
Year:	2016
Edition:	Single
Keywords:	Data mining, mortgage default, government sponsored enterprises, delinquency, credit score, loan-to-value, debt-to-income
Type:	Full Paper
First Page:	187
Last Page:	194
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Recent advances in information technology have made possible the analysis of vast amounts of data. One promising area for the application of the new analytical methods is finance. We perform data mining on the Fannie Mae mortgage portfolio from the fourth quarter of 2007 that includes 341,348 mortgages with the total principal value of more than $70 billion. This portfolio had the highest delinquency rate in the agency’s history – 19.4% versus the historical average of 1.7%. We find that although a number of information variables that were available at the time of mortgage acquisition in Q4, 2007 are correlated with the subsequent delinquencies, application of data mining techniques fails to accurately capture the mortgage delinquency patterns in the historical data. These results are consistent with an exogenous shock explanation and reveal a fundamental challenge that can arise in data mining large datasets.
	Go Back