This project is a Reinforcement Learning-based derivative hedging platform designed for option pricing, portfolio hedging, and risk simulation. It features an interactive Streamlit frontend and a Python backend that uses RL models (PPO, LSTM, Transformer) for decision-making in simulated market environments. This tool is meant for educational and simulation purposes only. It is not financial advice.
Features include:
- RL Policies: PPO, LSTM-enhanced PPO, Transformer RL
- Option Pricing: Black-Scholes baseline + RL overlay
- Portfolio Hedging: Simulated multiple options portfolio
- Greeks: Delta, Gamma, Vega
- Risk Metrics: PnL and basic VaR simulation
- Export: PDF reports for hedging and portfolio results
- Interactive Streamlit dashboard with tabs and plots
- Real-time market data fetching via Yahoo Finance
Handles dashboard, user input, visualizations, PDF exports
Handles RL training, model inference, market simulation
https://rl-hedging-system-r5gjbev2aymrdupzq6sref.streamlit.app/#rl-based-derivative-hedging
This project is for simulation and educational purposes only. Not intended for live financial trading. Use at your own risk.