This is a personal project to demonstrate whether Federal Funds rate changes can be accurately predicted using FOMC meetings minutes, by leveraging machine learning, primarily NLP (Natural Language Processing). The FOMC is the policy governing body of the US Federal Reserve Bank which is the central bank of the US.
The results achieved were a 94% prediction rate on test data (unseen data). I used four different machine learning models: Naive Bayes, Logistic Regression, Support Vector Machine (SVM), and Decision Trees. The model that worked best was Naive Bayes.
To see a presentation of this project online click here.
For a detailed wriiten overview of this project see the pdf in this repo called "Fed_Funds_ML_Project_Report". This report includes sections on project overview, data wrangling, statistical analysis, and machine learning.
Use this on the command line to start your Jupyter Notebook to get more processing power else it will stall out:
jupyter notebook --NotebookApp.iopub_data_rate_limit=10000000000
This was built by Michelle Bonat. It's not currently open for contributions, but I would love to hear any comments and suggestions about how you have modified this code. Contact me through the methods noted below.
- Michelle Bonat - Initial work - Contact me through michellebonat.com or on GitHub
This project is licensed under the MIT License - see the LICENSE.md file for details