Financial-Risk-Modelling

Overview

This project focuses on financial risk modeling using credit data from the Bank of Baroda. The primary objective is to analyze internal product files and CIBIL reports, clean and preprocess the data, and apply machine learning techniques to assess credit risk.

Dataset

• Case Study 1: Internal product file (Bank of Baroda)
• Case Study 2: CIBIL report for the same dataset

Steps Involved

1. Data Preprocessing

• Uploading and loading the datasets
• Cleaning missing values (e.g., removing rows with excessive null values)
• Merging datasets based on common parameters
• Handling imbalanced data

2. Exploratory Data Analysis (EDA)

• Statistical summary of features
• Visualizations to identify trends and correlations
• Checking for multicollinearity using Variance Inflation Factor (VIF)

3. Machine Learning Model

• Splitting the dataset into training and testing sets
• Implementing a Random Forest Classifier
• Evaluating performance using:
  • Accuracy Score
  • Precision, Recall, and F1-score
  • Classification Report

Dependencies

Ensure you have the following Python libraries installed:

pip install numpy pandas matplotlib scikit-learn statsmodels

How to Run

1. Clone the repository:

   git clone https://github.com/your-repo/Financial-Risk-Modelling.git

2. Navigate to the project directory:

   cd Financial-Risk-Modelling

3. Run the Jupyter Notebook:

   jupyter notebook Credit_Modelling_Project_BOB.ipynb

Results

• The model's predictions on credit risk classification
• Insights derived from data analysis
• Potential areas for further optimization

Future Enhancements

• Implementing additional ML models (Logistic Regression, XGBoost, etc.)
• Enhancing feature engineering
• Deploying the model as a web application

Contributors

• Your Name

License

This project is licensed under the MIT License.