Research Build

Stock Market Volatility Forecasting

Built a financial prediction pipeline that combines classical volatility modeling with neural forecasting methods for short-term risk estimation.

Year2024

Impact

Compared traditional and hybrid forecasting methods for better volatility and Value-at-Risk estimation under market uncertainty.

Problem

Financial volatility estimation needs both statistical rigor and adaptability to changing temporal dynamics, especially for short-term risk forecasting.

Approach

I combined GARCH-based modeling with LSTM forecasting to compare classical and hybrid approaches on historical S&P 500 volatility behavior.

Outcome

The project produced a stronger understanding of how hybrid financial forecasting models can improve volatility estimation and Value-at-Risk analysis.