Testing Methodologies for Credit Score Models to Identify Statistical Bias toward Protected Classes
This white paper provides lenders with a statistical methodology for determining whether the credit score model in use favors one population over another. The risk of such outcomes could constitute concerns related to the Equal Credit Opportunity Act (ECOA).
The paper discusses how to appropriately analyze and measure evidence of statistical bias in a credit score model that can impact a lender’s credit extension and underwriting policies. The paper presents insights on statistical bias in both the secured and unsecured lending industries, and under recessionary and non-recessionary environments.
Using this methodology, lenders can test outcomes with respect to sub-populations, including race or ethnicity, national origin, marital status, age and religion, among other demographics, across product types to determine if there is a statistically significant difference in probability of default.
Related Downloads:
Popular Articles
Consumer FAQ: Benefits of Adding Rent and Utility Data to a Credit File
Advantage of Adding Rent and Utility Data whitepaper
Credit with a Conscience fact sheet
Driving Financial Inclusion with Data and Analytics fact sheet