Skip to content

A Deep Reinforcement Learning neural net for an original Multi-Dimensional Pairs Trading strategy is proposed

Notifications You must be signed in to change notification settings

tiagomonteirocardoso/Deep-Reinforcement-Learning-in-MultiDimensional-Pairs-Trading

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

Deep Reinforcement Learning for Multi-Dimensional Pairs Trading

This project consists of a deep reinforcement learning agent which is able to conduct an automated trading strategy based on the new concept of multi-dimensional pairs trading. The folder DRLMPT contains all source code plus the folder containing the data.

Installation

The project is implemented in python 3.5.5. Apart from python version 3, the following libraries must be installed:

  • theano
  • lasagne
  • numpy
  • statsmodels
  • pandas

All these libraries can be easily installed through Anaconda, a python package management tool free available for download.

After downloading and installing Anaconda, you can install the required libraries easily with the following command:

conda install theano lasagne numpy statsmodels pandas

Running

To run DRLMPT, execute runner.py in the DRLMPT folder.

cd DRLMPT
python3 runner

About

A Deep Reinforcement Learning neural net for an original Multi-Dimensional Pairs Trading strategy is proposed

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages