This project is my first step to creating a trading robot that uses a Gradient boosting classifier to not only make predictions about a trade but to also execute the trade on live data. This model is tested against past data with the currency pair being USD/CAD. The symbols can be easily changed for generalization.
- The csv files were extracted from Dukascopy.
- The forex that we try to predict here is USD/CAD.
- All datetime indexes are in GMT.
To run this project, you'll need the following enviroments and libraries:
- Python 3.X
- Jupyter Notebook
- Numpy
- Pandas
- Scipy
- Sklearn
- Matplotlib
This example uses sample data from December 2021. 95% of the data is used for training and the remaining 5% for testing.
Green and red vertical lines represent winning trade and losing trade respectively.
