Home

DQN Trading Simulation

Exploration of trading strategies using Deep Q-Networks (DQN),
applying reinforcement learning to historical ETF price data (1655.T).

---

📚 Contents

• How to Set Up
• Project Overview
• Data Preprocessing
• Training Method
• Evaluation Method
• Future Work

---

🚀 About This Project

• Focus: Applying DQN (Deep Q-Networks) to ETF (1655.T) trading simulations
• Language: Python
• Libraries: TensorFlow, scikit-learn, yfinance, pandas, matplotlib, numpy
• Dataset: Yahoo Finance (1655.T) from 2017-09-29 to 2023-03-31
• License: MIT License

---

🔥 Future Plans

• Add advanced evaluation metrics
• Explore improvements with Double DQN or Dueling DQN
• Expand to more financial instruments
• Create detailed documentation and tutorials

---

📝 Notes

This project is for research and educational purposes only.
It is not intended as financial advice.