End-to-end Credit Risk Scorecard Development in Python & SQL. Features Weight of Evidence (WOE) / Information Value (IV) analysis, Logistic Regression modeling, and PD risk segmentation.
-
Updated
Aug 17, 2026 - Python
#
woe-iv
Here are 7 public repositories matching this topic...
Consumer credit risk, PD scorecard (logistic regression + WOE/IV) on Home Credit data, with discrimination, calibration and population-stability validation, ongoing monitoring, and model governance. Includes an EAD analysis reframed as a documented data-quality finding.
logistic-regression ead model-validation credit-risk scikit-learn-python model-monitoring credit-scorecard pd-model model-governance consumer-credit woe-iv
-
Updated
Jun 23, 2026 - Jupyter Notebook
Credit risk scoring model using Logistic Regression, Decision Tree, and XGBoost (Give Me Some Credit dataset)
python finance data-science machine-learning banking xgboost logistic-regression credit-risk woe-iv scorehard
-
Updated
Aug 13, 2026 - Jupyter Notebook
Interpretable insurance approval model: WoE/IV features, SMOTE, Elastic Net, AUROC 0.83 from a 1.5% base rate
-
Updated
Sep 5, 2026 - Jupyter Notebook
End-to-end consumer-credit scorecard — WOE binning, Information Value feature selection and logistic regression, scaled to a point-based scorecard and evaluated with KS/AUC (KS 0.52, AUC 0.83).
-
Updated
Sep 12, 2026 - Python
Config-driven, cross-platform ML pipeline for binary risk scoring (WoE/IV, scorecards, calibration, SHAP, ONNX, FastAPI serving, PSI/CSI monitoring).
machine-learning scikit-learn credit-scoring scorecard binary-classification shap fastapi model-monitoring risk-modeling woe-iv
-
Updated
Jun 11, 2026 - Python
Credit Risk Scoring: WOE/IV + XGBoost/LightGBM + SHAP | OOF AUC 0.785 · KS 0.430 | Home Credit Default Risk
python machine-learning data-analytics xgboost lightgbm feature-engineering credit-risk shap streamlit woe-iv
-
Updated
Sep 6, 2026 - Python
Add this topic to your repo
To associate your repository with the woe-iv topic, visit your repo's landing page and select "manage topics."