Deep Reinforcement learning and Trading

The goal of this project is to explore Reinforcement Learning algorithms for the use of designing systematic trading strategies on futures data. This could address most parts of the trading strategy lifecycle including signal extraction, portfolio construction and risk management. Special consideration will be given to the non-stationarity problem as well as limited data for model training purposes.

Direct Supervisor: Ali Hirsa

Position Dates: 6/1/2020 - 8/31/2020

Hours per Week: 20-40


Eligibility: SEAS only

Ali Hirsa, [email protected]