/stock-Q-learning

Tools for using Q-learning to train an Agent to trade stocks, relying on technical indicators as guidance.

Primary LanguagePython

stock-Q-learning

Motivation

This is a system for training a computer to trade stocks using Q-learning. The states that are fed to the Q-learning algorithm are based off of a combination of several technical indicators while the reward is the return the agent achieves the next day.

Requirements

You'll need Python 2.7 and the following libraries:

  • numpy
  • pandas
  • matplotlib