Credit card default risk scorecard

The brief

A bank wants an explainable scorecard to decide credit limits and early collections. Build a default-risk model from repayment history and bill amounts, and turn it into a points-based scorecard credit officers can read. Good looks like: good discrimination (AUC, KS), a readable scorecard, and fairness checks.

What you will build

1. Explore repayment behaviour and default rates
2. Build features from payment history
3. Train logistic regression and a gradient-boosting model
4. Convert the logistic model to a points scorecard
5. Check fairness by gender and age and write the policy note

Free, open data to build it with

• Default of Credit Card Clients ↗ UCI Machine Learning Repository

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.