Shafi2016 / Comparing-Out-of-Sample-Performance-of-Machine-Learning-Methods-to-Forecast-U.S.-GDP-Growth

Published in Computational Economics

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

About The Project

This repository provides Python codes to perform all the calculations in the paper : Comparing Out-of-Sample Performance of Machine Learning Methods to Forecast U.S. GDP Growth

Pre-requisites

List of Python Files

Filename Description
forecast_GDP_midas.py to forecast GDP growth with MIDAS
forecast_GDP_v2.py to forecast GDP growth with ML and DL methods
my_algorithms_v2.py to implement all the ML and DL algorithms described in the paper
forecast_GDP_v2.ipynb to do all data preprocessing and transformation and to forecast GDP growth using $H_2 O$
graphs_rev1.ipynb to plot all graphs in the paper

License

Distributed under the MIT License. See LICENSE.txt for more information.

Contact

Ba Chu - ba.chu@carleton.ca

Project Link: https://github.com/wave1122/horserace1

About

Published in Computational Economics

License:MIT License


Languages

Language:Jupyter Notebook 97.3%Language:Python 2.7%