https://github.com/caesaredia/credit-default-risk-analysis
Predicting customer loan default risk using exploratory data analysis and basic statistical modeling.
https://github.com/caesaredia/credit-default-risk-analysis
Last synced: 10 months ago
JSON representation
Predicting customer loan default risk using exploratory data analysis and basic statistical modeling.
- Host: GitHub
- URL: https://github.com/caesaredia/credit-default-risk-analysis
- Owner: caesaredia
- Created: 2025-04-09T08:15:46.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-04-10T05:45:28.000Z (over 1 year ago)
- Last Synced: 2025-05-07T20:18:16.354Z (about 1 year ago)
- Language: Jupyter Notebook
- Size: 587 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Credit Default Risk Analysis
This project analyzes factors that may influence the probability of customer loan default using a real-world dataset from a financial institution. The analysis focuses on identifying trends and relationships across demographic and financial variables.
## Project Objectives 📊
- Understand which customer attributes contribute to higher or lower loan default risk.
- Provide insights that can help credit divisions make more informed lending decisions.
- Apply data cleaning, transformation, and exploratory analysis techniques.
## Key Analysis & Hypotheses Tested 🔍
- **Number of Children**: Are customers with more children more likely to default?
- **Family Status**: How does marital status influence default probability?
- **Income Level**: Do low-income customers default more often?
- **Loan Purpose**: Are loans for education or car purchases riskier?
## Tools Used
- Python (pandas, matplotlib, seaborn)
- Jupyter Notebook
## Files
- `credit_analysis.ipynb` — the main notebook containing all analysis steps
- `credit_scoring_eng.csv` — source data file
## Notes
- Outliers and missing values were addressed using median imputation and basic filtering.
- The dataset was cleaned for inconsistencies like casing differences and invalid values (e.g., age = 0, children = -1).
## Author
Nabilla Hafsah Caesaredia