Credit Classification Using Grammatical Evolution
Paper: Credit Classification Using Grammatical Evolution
Anthony Brabazon and Michael O’Neill
Grammatical Evolution (GE) is a novel data driven, model induction tool, inspired by the biological geneto-protein mapping process. This study provides an introduction to GE, and demonstrates the methodology by applying it to model the corporate bond-issuer credit rating process, using information drawn from the financial statements of bond-issuing firms. Financial data and the associated Standard & Poor’s issuer credit ratings of 791 US firms, drawn from the year 1999/2000 are used to train and test the model. The best developed model was found to be able to discriminate in-sample (out-of-sample) between investment grade and junk bond ratings with an average accuracy of 87.59 (84.92)% across a five-fold cross validation.
Related book:
Biologically Inspired Algorithms for Financial Modelling
April 29th, 2008 at 7:35 am
Color me unimpressed… 791 training points? How many free parameters does this “non parametric” model contain?
A real benchmark for these methods is something like the netflix dataset with >10^8 training points and 2 million test points..
April 29th, 2008 at 7:57 am
You have a point, or half a point ;-)
Take the book for what it is: “provides an introduction to a broad range of biologically inspired algorithms and illustrates how they can be applied to financial modelling…etc…..we hope it will help spark new ideas …etc..”