AlgoTradingML

Comparison of Different Algorithmic Trading Strategies on Tesla Stock Price

Tawfiq Jawhar

Machine Learning (COMP-652 and ECSE-608)
Fall 2018
McGill University

Instructors:
Audrey Durand
Riashat Islam

Notebooks

1- Trading Environment and Buy and Hold Benchmark

2- Extracting Data for Learning

3- Exploring TSLA

4- Regime Detection with GMM

5- SMA Optimization

6- SE-MA Optimization

7- MA with GMM Risk Manager

8- Price Movement Classifiers

Installation

The trading environment will be used for backtesting is Zipline which is an open source Python algorithmic trading library. Zipline is used as the backtesting and live-trading engine powering Quantopian.

To simplify the installation I created a docker image on DockerHub. The docker image is based on Dockerfile found in zipline repo.

To pull the docker image:

docker pull tawjaw/ziplineml

To run the docker use:

docker run -v /path/to/your/notebooks:/projects -v ~/.zipline:/root/.zipline -p 8888:8888/tcp --name zipline -it docker pull tawjaw/ziplineml:v1

The default password for Jupyter is jupyter.