Mini-project
Banking & Fintech
Loan approval fairness study
Test with statistics whether loan approvals differ across groups of applicants.
The brief
A lender's compliance team must show that its approval decisions are fair. Use public lending data to test whether approval rates differ across groups of applicants once income and loan size are taken into account, and explain what the numbers do and do not prove. Good looks like: correct tests, adjusted comparisons, honest limits, and charts a compliance officer can present.
What you will build
- Explore approval rates by group
- Test raw differences with chi-square tests and confidence intervals
- Adjust for income and loan size with logistic regression
- Discuss correlation versus causation and missing factors
- Present the findings in a short report
Free, open data to build it with
- HMDA mortgage data browser ↗ Consumer Financial Protection Bureau
- Statlog (German Credit Data) ↗ 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.