The purpose of this project is to gain hands-on experience with time series data.
- Data Visualization
- Inferential Statistics
- Time Series
-Python
This project started as a simple attempt to re-write an older project from EViews to Python. But, it quickly took on its own form. After making a couple of small changes to the data set, I realized that it was a great opportunity to gain some experience with something I hadn't worked with before, namely time series modeling. And, so, I decided to pursue this as an experiment rather than a solution to some problem. As such, I was not worried about data leakage. This afforded me the opportunity to play with the models, and to analyze the differences between them while all other things remain constant. This is Part 1 of 2.
The data are from the St. Louis Federal Reserve Economic Data site and Yahoo Finance.
The models used, here, in Part 1 are Linear Regression, Principal Components Regression, and Autoregressive Integrated Moving Average (ARIMA).
The paper describes the results from the original project.
- Clone this repo.
- Raw data are kept in /assets