Trading strategy of High Frequency Data using ML (Classification)

The ultimate goal of this project is to implement an end-to-end trading strategy using machine learning. This deliverable contains the modeling part of the final project of WQU Machine Learning in Finance Course (check the program for more details here https://wqu.org). The project will apply an investment strategy using classification.

• README: this file
• model.ipynb: notebook of the model solution with thorough discussion and conclusions

S&P500 E-Mini tick data

All the results are discussed in the notebook

This python script has been written to be run using Python 3.6. TA-Lib: see http://ta-lib.org/ for instructions on the installation

The key modules required for this project are the following:

re 2.2.1 json 2.0.9 pandas 0.22.0 pandas_datareader 0.7.0 numpy 1.15.2 sklearn 0.19.1 statsmodels 0.8.0 scipy 1.0.1 matplotlib 2.2.2 seaborn 0.8.1 talib 0.4.17

Most of the above key modules can be installed in Anaconda except ta-lib (look at the project's webpage) The cleanest way to proceed would be to create an environment with the required modules. See more info on how to create environments here: https://docs.python.org/3/tutorial/venv.html All the libraries can be installed using pip within the env:

pip install

alvaro calle cordon - Data Scientist -