-
This repository is a collection of the projects I completed for Economics 144 at UCLA with Dr. Randall Rojas, for Spring 2023.
-
While the formal class title is "Economic Forecasting," the course covers time series analysis, modeling of time series data, and forecasting, with applications across many disciplines, not just economics.
-
There were three projects, each with increasing complexity. Thus the third project is most representative of the skills I learned.
-
Some examples of models we studied are the following:
- ARIMA, ETS, Prophet (developed by Facebook), Neural Network Autoregression, State Space Models (e.g. Kalman Filtering), Holt-Winters, GARCH (for conditional variance), VAR, Cointegration, THIEF, MAPA, Dynamic Harmonic Regression, STL decomposition, TBATS, model combinations
- Most (but not all) of the aforementioned models were used. The third project in particular incorporates most of these elements.
-
For projects 2 and 3, I decided that I disliked the look of the standard graphing functions in the libraries I was using, so I created my own visualizations from scratch using
ggplot2. -
All work is my own (i.e. analyses, models, visualizations), and it was all done in R. I also found the data sets myself.
Each folder has its own README, within which I go into more detail for the components of each respective project.