Glossary of Credit Scoring Terms
Here are definitions related to credit scoring: Credit Bureau Ratings - one of four types of credit scores. These include the D&B Paydex and Experian's DBT. These are non-statistical scores you buy from a credit bureau. Credit Scorecards - Are heuristic (knowledge-based) and usually reflect the factors that a credit analyst might incorporate in making a credit decision. Scorecards can include both internal and external data. Many bolt-on credit and collection software products contain a credit scorecard tool. Custom score - (also referred to as a "custom scoring model." Statistical models are often described as a scorecard, a pooled scorecard, and a custom scorecard. A scorecard uses data from one firm. A pooled scorecard uses data from many firms. A custom scorecard blends a statistical model with some of the factors used in a judgmental model. (Thanks to John Sargent of CMS) Generic scores - One of the four types of business-to-business credit scores. Generic scoring models are statistically derived from either a bureau's database such as Experian's IntelliscorePlus and D&B's Commercial Credit Score. They can also be generated by a third party model plugged into your customer database such as with Predictive Metrics' Net30Score. Hold-out Sample - A small percentage of your sample (typically 20 percent) is excluded from your model development process so that it can be used to test the final scorecard. See also validation sample. Lift - Something that raises, or "lifts" the predictive capability of your model. Validation Sample - A small percentage of your sample (typically 20 percent) is excluded from your model development process so that it can be used to test the final scorecard. Also known as a Hold-out Sample. Related material:
See our department on Credit Scoring.
| 
Webinars
