Validating Consumer Credit Risk Models

What Methods Are Relevant to Assuring Your Model's Stability in a Volatile Business Environment

Publication Date	April 2009
Publisher	IDC
Product Type	Report
Pages	18
ISBN Number	not applicable
Product Code	IDC06866

Buy this product or for assistance call +44 20 7060 7474

PDF £2,705.00$4500 | €3028

FAX Order above formats by FAX

Summary

This Financial Insights report details why a validation is essential, the appropriate steps to carry out a validation analysis, and who the business constituents are that rely on validations and also details what we mean by model "performance."

"As the global economy struggles in recession, lenders across industry verticals should closely track and validate their scoring models. During economic times of rapid expansion and contraction, scoring models can degrade in their ability to make optimal loan recommendations. Sound population and scorecard tracking and validation practices are part of a robust enterprise risk management program." - Dana Wiklund, research director, Global Risk Management

Content

Financial Insights Opinion
In This Report
Situation Overview
- Variability Within a Population
- Consumer Behavior
- Who Uses Validations?
- Who Should Perform Validation Analytics?
- How Validations Fit Within an Enterprise Risk Management Context
Method Specifics
- Tracking Through-the-Door Populations
  - Final Score Report
  - Figure: Final Score Report
  - Population Stability
  - Figure: Population Stability
  - Characteristic Analysis
  - Figure: Characteristic Level Changes
- Interval Bad Rates
- Figure: Interval Bad Rate Analysis
- Validation Sample
- Components of a Score Validation
- Statistical Measures Used in Score Validations
  - Kolmogorov-Smirnov Statistic
  - Figure: Kolmogorov-Smirnov Statistic
  - Divergence
  - Figure: Divergence
  - Predictiveness Index
  - Figure: Predictiveness Index
  - Information Value
  - Figure: Information Value
Future Outlook
Essential Guidance
- Actions to Consider
Learn More
- Related Research
- Synopsis