Financial Engineer

In this project, I explore the challenging task of forecasting stock returns using machine learning, combining feature selection and deep learning techniques. By integrating multiple selection methods with a neural network model, I aim to extract meaningful signals from noisy financial data and test the potential of supervised learning in stock prediction. 

This project represents my first exploration of copulas in statistical arbitrage through a pair trading strategy. By capturing complex dependence structures beyond simple correlation, copulas provide a flexible framework to model non-linear co-movements essential for effective trading signal generation.