This projects leverages Generative Adversarial Networks (GANs) to predict and generate realistic volatility surfaces for option pricing. It aims to enhance derivatives trading strategies by providing more accurate models for volatility, helping traders optimize hedging and identify arbitrage opportunities.
dataset used : https://www.cboe.com/tradable_products/vix/vix_historical_data/ Vix Index data from 2004 to present
- Realistic Volatility Surfaces: Generated using advanced GANs and VAEs. 8 Derivatives Trading Optimization: Improve pricing and hedging strategies with precise predictions.
- Stochastic Volatility Models: Integrated financial models (e.g., Heston, SABR) for added flexibility.
- Data-Driven: Train and evaluate using historical market data.
- Python, TensorFlow/PyTorch, QuantLib, Pandas, NumPy
- GANs, VAEs, Stochastic Volatility Models
- Matplotlib, Plotly for visualization
- Option Pricing: Enhance pricing strategies with better volatility forecasts.
- Arbitrage: Identify mispriced options in the market.
- Hedging: Improve risk management with more precise volatility surfaces.