AutoTrading with reinforcement learning

Abstract

[en] Trading is the act of studying any financial market and making money with it through buying and selling of assets. In this project, I will try to automate the actions performed by a trader without having a thorough knowledge of the financial market or trading techniques. I will use algorithms based on reinforcement learning techniques used in other fields such as robotics without human interaction in the algorithm’s execution. The main objective of this project is to investigate the feasibility of using these techniques adapted to Deep Learning and the ability to cope with the volatility of cryptocurrency. Furthermore, to show the results of these algorithms, the cryptocurrency Bitcoin and ADA will be used as a market study by obtaining the historical and making market analysis.