Data defines the model by dint of genetic programming, producing the best decile table.


Credit Risk Modeling – A Machine Learning Approach
Bruce Ratner, Ph.D.

With increasing delinquencies and credit losses on a national scale, creditors are taking a total view of credit risk from a portfolio level. This is in stark contrast to current credit risk management at the loan level by setting loan-level credit criteria. Creditors are using statistical regression modeling to predict future loan losses at the portfolio-level. However, the statistical regression approach is proving ineffective, because the regression paradigm of “fitting the data” to a pre-specified model is not tenable. The credit-risk data captures fluctuating economic conditions and ever-changing loan characteristics that makes fitting the data virtually impossible. The purpose of the article is to present the GenIQ Model© that lets the “data specify the model,” regardless of the data’s fluctuating conditions and ever-changing characteristics. GenIQ provides forecasted portfolio losses, which goes beyond the limited risk score.

For more information about this article, call me at 516.791.3544, or e-mail, br@dmstat1.com.
My publisher owns the copyright of the article, about which this abstract addresses. The article will appear in my forthcoming book.
My publisher has granted me permission to discuss orally the article's content, but by no means provide an outline, a draft or proof-ready of the article.


Sign-up for a free GenIQ webcast: Click here.