Investigating Algorithmic Stock Market Trading using Efficient Ensemble Techniques

This is an assisting repository for the published paper investigating ensemble methods in algorithmic trading. It is currently pending peer review. It was written by Khaled Sharif and Mohammad Abu-Ghazaleh, and was supervised by Dr Ramzi Saifan.

Recent advances in the machine learning field have given rise to efficient ensemble methods that accurately forecast time-series. In this paper, we will use the Quantopian algorithmic stock market trading simulator to assess ensemble method performance in daily prediction and trading; simulation results show significant returns relative to the benchmark and strengthen the role of machine learning in stock market trading.