dukererenst/oanda_trading

Trading in the deep.

Automatic Live Trading with the Oanda API Based on Python Multithreading

Oanda_trading is an open-sourced software written in Python to perform algorimic trading with Oanda API. We apply the multithreading framework to handle the market data streaming, trading signal calculation, and order sending.

Getting Started

1. Obtain the access token and account ID from the official website of OANDA.
2. Install MongoDB(v3.2.4), start it as system service.
3. Install Python(2.7.13) , PyMongo(v3.2.2) and other necessary libraries.
4. After putting the python scripts in the same directory, modify the access token and account ID in settings.py, change the logging setting in trading_log.py.
5. Start MongoDB as system service:

sudo service mongod start

6. Create the database and collections mentioned in the main script(forex_trading_general_171005.py) in MongoDB.
7. Open a terminal enter the same directory of these python scripts, and start to run the trading program as below,

python forex_trading_general_171005.py

8. Now you can check the trading signal logs in the log file, and implement your own trading strategy in the main script.

Prerequisites

It is recommended that you run the script in a Ubuntu 14.04/16.04/17.04 LTS system where MognoDB is installed.

You also need an installation of Python 2.7 from Anaconda.

List of Python packages needed:

• Pandas
• Numpy
• PyMongo
• PyWavelets
• oandapy (Python wrapper for the OANDA REST API )
• python-requests

Installing

No installation is required so far for this trading library.

---

Deployment

Please read this blog (https://www.quantstart.com/articles/Forex-Trading-Diary-1-Automated-Forex-Trading-with-the-OANDA-API) on QuantStart.com before the deployment of this trading program on the clouds or your local servers

Documentation

Please check out architecture_of_oanda_trading.pdf for the multi-threading architecture of this trading system.

Gona make the Wiki page soon:)

Authors

• Jiansen Zheng - LinkedIn

[email protected]

License

This project is licensed under the GNU General Public License v3.0 - see the LICENSE.md file for details

Acknowledgments

• Inspired by Michael Halls-Moore from quantstart.com

Bitcoin Donation

• If you would like to support my project, please consider donating to this address: 35TkyWD4Vwp66hX8b8U8stABSDyW3pcCS1