Forecasting mortgage defaults: evidence from UK portfolio-level data
by Alexios Makropoulos; Anastasios Savvopoulos; Peter L. Jones
International Journal of Computational Economics and Econometrics (IJCEE), Vol. 5, No. 2, 2015

Abstract: This paper considers an equilibrium/error correction modelling (ECM) approach to identify determinants of mortgage portfolio default rates at firm level. Using mortgage portfolio data from a UK lender we estimated the lender's portfolio concentration weights in different UK regions to estimate portfolio-adjusted unemployment rate and portfolio-adjusted house price index. The modelling results suggest that the portfolio-adjusted unemployment rate and house price index along with Council of Mortgage Lenders' (CML) default rates, interest rates and household savings are useful determinants of firm-level default rates within an ECM framework. These findings may be of practical use for forecasting or back-populating default rates in financial institutions and especially for the smaller lenders where a long time-series with default rates may not be available. However, modelling and forecasting at portfolio level should be made carefully, as we argue that data quality and changes in the lending policy should be carefully monitored.

Online publication date: Wed, 08-Apr-2015

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Economics and Econometrics (IJCEE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com