Credit line increase eligibility
Proactive credit line increase decisions
Owner: Credit Risk Analytics · last validated Aug 29, 2026 · metrics illustrative on synthetic data
Outcome it answers for
Illustrative. In production this figure comes from the monitored champion-challenger result, not the card.
Top drivers (SHAP)
Disaggregated performance
| Group | AUC | Approval |
|---|---|---|
| Proxy group A | 0.803 | 41.2% |
| Proxy group B | 0.797 | 38.8% |
| Proxy group C | 0.799 | 37.6% |
| Proxy group D | 0.806 | 40.1% |
Groups are BISG race and ethnicity proxy groups; fair-lending testing applies because this is a credit decision. Synthetic.
Intended use
Ranks existing cardholders for proactive credit line increases. A credit decision: when a requested increase is declined, adverse-action rules apply.
Training data
24 months of account performance. Synthetic in this environment; metrics are illustrative.
Limitations
Cardholders under 6 months tenure excluded; recalibrated quarterly.
Fair lending and adverse action
Credit decision under ECOA and Reg B: declined requests receive specific principal reasons (12 CFR 1002.9). Fair-lending testing uses the adverse impact ratio on a BISG race and ethnicity proxy; an AIR below 0.80 triggers a search for less discriminatory alternatives. Proxy results here are synthetic.