This project provides a flexible and universal financial cluster analysis script using Torque Clustering. The script is designed to work with any financial time series data in CSV format, allowing you to identify patterns, clusters, and anomalies in price changes over time. The code is written in Python and includes data loading, clustering, and visualization components, making it a comprehensive starting point for financial data analysis.
🚀 Features
Flexible data loading from CSV files
Price change calculation (percentage)
Torque clustering for identifying market phases
Clear and customizable data visualizations
Easy to extend and integrate with other financial models
🔧 Installation
Make sure you have the required Python packages installed:
pip install pandas matplotlib numpy
📈 Sample Output
The script will produce a scatter plot showing price changes over time, with clusters color-coded for easier interpretation. It will also print a statistical summary of each cluster, including count, mean, standard deviation, minimum, and maximum values.
📁 Data Format
Your CSV file should have the following columns:
Open Time (datetime)
Close (float)
High (float)
Low (float)
Volume (float)
Example:
Open Time,Close,High,Low,Volume 2023-01-01 00:00:00,1200.5,1250.0,1190.0,3000 2023-01-01 12:00:00,1300.5,1350.0,1280.0,3500
📄 License
This project is licensed under the MIT License. Feel free to modify and distribute.
🤝 Contributing
Contributions are welcome! If you have ideas for improving this project or want to add new features, feel free to open a pull request or create an issue.