Mary Chu (SAS Institute Ltd)

Title: Data Mining For Credit Scoring Models with Enterprise Miner

Abstract: Credit today is granted by various organizations such as banks, 
retailers, utilities and others. Credit scores, both of prospects and 
existing customers, are essential in the customization of 
credit products. They are used during application stage and for 
determining custom credit limits, down payments or deposits and interest rates.

There are two types of modeling with credit scoring: Application and
Behavioral. Application models (or scorecards as they are commonly known) 
are need during the time of application whereas. Behavioral scorecards are 
applied to existing customers periodically to predict the 
likelihood that an account will go bad.

SAS Enterprise Miner software is used to build credit scoring models for 
the credit industry. A case study on behavioral scoring will be given. 
The development process will include data 
access, variable transformation and selection, model building and 
validation.