This is a submission for my Econometrics graduate course: Regression with Time-Series Data using STATA. There are 4 files: Time Series Problem with all the questions and requirements, copper.dta providing the data, Time Series (TXT) with STATA code, and Time Series Write Up with the result and detailed explanation.
Time series models are always difficult to deal with, and there are a lot of tests needed to perform before arriving to the most optimal regression model. In this problem set, I broke down the process into small steps in order to:
- Explain the differences between stationary and nonstationary time-series process
- Describe an autoregressive process and a random walk process
- Perform Dickey-Fuller and augmented Dickey-Fuller tests for stationary
- Explain cointegration
- Choose an appropriate model for regression analysis with time-series data