Volatility Forecasting in Financial Time Series

This project explores volatility forecasting for the NIFTY 50 index using financial time series analysis.

Contents

• volatility_forecasting.ipynb — full analysis notebook
• requirements.txt — Python dependencies
• generated charts: 01_nifty50_closing_price.png, 02_log_returns.png, etc.

Key steps

1. Collect or simulate NIFTY 50 data.
2. Compute log returns and rolling volatility (SMA / EWMA).
3. Test stationarity using ADF.
4. Build ARIMA and machine learning models.
5. Compare model performance with MAE and RMSE.

How to run

pip install -r requirements.txt
jupyter notebook volatility_forecasting.ipynb

Notes

• The notebook uses synthetic data when internet access is not available.
• The venv folder is excluded from Git.