Skip to content

Latest commit

 

History

33 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Banking Credit Risk Modeling and Scorecard Development

End-to-end credit risk modeling framework covering Probability of Default (PD), Loss Given Default (LGD), Exposure at Default (EAD), credit scorecard development, and Expected Loss estimation using Python and Basel II risk management principles.

Overview

This project demonstrates the complete credit risk modeling lifecycle used in banking and consumer lending. The workflow includes data preparation, risk modeling, scorecard development, model validation, portfolio monitoring, and Expected Loss estimation.

Business Problem

Financial institutions need reliable methods to assess borrower risk, estimate potential losses, and support data-driven lending decisions while maintaining regulatory compliance.

Key Objectives

  • Build Probability of Default (PD) models
  • Develop Loss Given Default (LGD) models
  • Develop Exposure at Default (EAD) models
  • Create credit scorecards
  • Calculate Expected Loss (EL)
  • Monitor model stability and performance

Dataset

Consumer lending dataset containing borrower demographics, credit history, loan characteristics, repayment behavior, and default outcomes.

Methodology

Data Preparation

  • Data cleaning
  • Missing value treatment
  • Feature engineering

Risk Modeling

  • Weight of Evidence (WoE)
  • Information Value (IV)
  • Logistic Regression
  • PD Modeling
  • LGD Modeling
  • EAD Modeling

Validation & Monitoring

  • ROC-AUC
  • Gini Coefficient
  • KS Statistic
  • Population Stability Index (PSI)

Tools & Technologies

  • Python
  • Pandas
  • NumPy
  • Scikit-learn
  • Matplotlib
  • Seaborn
  • Jupyter Notebook

Key Concepts

  • Credit Risk Modeling
  • Basel II
  • Probability of Default (PD)
  • Loss Given Default (LGD)
  • Exposure at Default (EAD)
  • Expected Loss (EL)
  • Credit Scorecards
  • Weight of Evidence (WoE)
  • Information Value (IV)
  • Population Stability Index (PSI)

My Contributions

  • Data preprocessing and feature engineering
  • Exploratory data analysis
  • WoE and IV analysis
  • PD, LGD, and EAD model development
  • Credit scorecard construction
  • Model validation and monitoring
  • Documentation and project organization

Results

Successfully developed an end-to-end credit risk framework capable of estimating borrower default risk, potential losses, and overall portfolio exposure.

Future Enhancements

  • XGBoost and LightGBM models
  • MLflow experiment tracking
  • FastAPI deployment
  • Power BI reporting dashboards
  • Cloud deployment

Repository Structure

credit_risk_modelling_python/
│
├── data/
├── notebooks/
├── models/
├── reports/
├── README.md
└── requirements.txt

Disclaimer

This project was completed as part of the 365 Data Science Credit Risk Modeling in Python course. All code implementation, analysis, model development, and repository documentation were completed by me for learning and portfolio purposes.

License

MIT License

About

Completed as part of the 365 Data Science Credit Risk Modeling in Python Udemy course. Developed an end-to-end credit risk modeling pipeline for consumer lending, covering data preprocessing, feature engineering, Probability of Default , Loss Given Default , Exposure at Default , scorecard development, model validation, population stability

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages