Algorithm Trading using Q-Learning implemented on Quantopian Zipline platform (https://github.com/quantopian/zipline).
In this python code, I will present an reinforcement learning framework to trade a single stock. This area of machine learning consists in training an agent by reward and punishment without needing to specify the expected action. The agent learns from its experience and develops a strategy that maximizes its profits. The stock daily states are based off of a combination of several technical indicators (RSI, Bollinger, Momentum, et al).
T.M. Mitchell. Machine Learning. McGraw-Hill International Editions, 1997.
This project requires Python 2.7 and the following Python libraries installed:
- Matplotlib
- NumPy
- Pandas
- Seaborn
- Run
- Zipline