UK retail banks – commission report

UK news story: Retail banks should be ring-fenced, says Independent Commission on Banking:
http://www.bbc.co.uk/news/business-13032403

Loss given default models incorporating macroeconomic variables for credit cards

New paper accepted for publication:

• T. Bellotti and J. Crook. Loss given default models incorporating macroeconomic variables for credit cards, International Journal of Forecasting, Article in Press 2011, Accepted Manuscript (available online).

Abstract

Loss Given Default is an important measure of credit loss used by financial institutions to compute risk within credit portfolios, expected loss on individual loans and capital requirements.  The Basel II Capital Accord gives banks the opportunity to calculate their own estimates of Loss Given Default.  Based on UK data for major retail credit cards, we build several models of Loss Given Default based on account level data including Tobit, a decision tree model, a Beta and fractional logit transformation.  However, we find that Ordinary Least Squares models with macroeconomic variables perform best for forecast of Loss Given Default at account and portfolio level on independent hold-out data sets.  The inclusion of macroeconomic conditions in the model is important since it provides a means to model Loss Given Default in downturn conditions as required by Basel II and enables stress testing. 

Center for Operations Research and Econometrics

CORE – ECORE

 

Econometrics seminar/LSM Finance seminar

 

 

Speaker:

Tony BELLOTTI, Imperial College London

 

Title:

Support vector machines in finance: application to the prediction of bank ratings

 

Date and place:

Wednesday, December 8,  2010 at 14h30 

CORE – room b_135

Abstract: 

We compare the ability of ordinal choice models and support vector machines to model and predict international bank ratings. Although support vector machines can identify the significant determinants of ratings we argue that ordered choice models are more reliable for this purpose. Our findings suggest that ratings reflect a bank’s financial position, the timing of when the rating was made and a bank’s country of origin. Accounting for country effects in the model was found to be particularly important not least because they substantially improve the predictive performance of the models. We find that support vector machines can produce considerably better in- sample predictions of international bank ratings than the standard method currently used for this purpose, ordered choice models. This appears to be due to the support vector machine’s ability to estimate a large number of country dummies unrestrictedly, which was not possible with the ordered choice models due to the small sample size. Given that the primary purpose of modelling ratings is prediction this is an important result.

(This is joint work with R. Matousek and C. Stewart.)

A paper is available for consultation – Please consult web page:

http://www.uclouvain.be/en-43677.html