• Conducted a volatility study to develop pairs trading strategy by writing web crawlers that automated extracting 30 equity and ETF spot and options prices data from CBOE and Yahoo Finance • Utilized NumPy, Pandas, and SciPy packages to calculate implied volatility, realized volatility, and risk premiums to measure how the market prices risk • Gathered and plotted daily VIX futures data with Selenium, Seaborn and Matplotlib to study volatility term structure • Examined volatility clustering and built forecasting tools for market risk using correlations of daily absolute returns and volatility at different time lags

Forecast next-month S&P 500 equity premium using classic macro/valuation predictors and compare econometric vs. ML models under consistent rolling/expanding backtests; evaluate with OOS 𝑅 2 R 2 and trading Sharpe; includes an IBM single-stock extension.

Replicate Neely, Rapach, Zhou, 2014, Forecasting the Equity Risk Premium

Replication of the results of Forecasting Equity Premium: The Role of Technical Indicators (Neely(2014))

Forecasting the equity premium using ordinary LS approach, using ML algorithms with comparison and conclusion

Empirical forecasting of the equity risk premium using historical financial predictors and time-series regression methods.

Provides the code and data for the paper "Forecasting the equity premium-Can machine learning beat the historical average"

This study demonstrates that machine learning models—especially Random Forests, SVMs, and Neural Networks—can outperform traditional models like CAPM in forecasting the equity risk premium in terms of out-of-sample R²