Week 9: ARIMA models
What you will learn this week
- Seasonal ARIMA models
- Computing forecasts for ARIMA models
- ARIMA vs ETS models
Pre-seminar activities
Read Sections 9.8-9.10 of the textbook and watch all embedded videos
Tutorial exercises
NoteTutorial Learning Objectives
- Develop and evaluate seasonal ARIMA models.
- Able to transform/difference seasonal series, fit and diagnose competing ARIMA models, and produce forecasts (including comparing against ETS and STL-adjusted approaches).
- Compute ARIMA-based forecasts manually.
- Able to use a specified ARIMA model equation and estimated parameters to hand-calculate multi-step forecasts, then fit the same model in R and explain differences between manual and software forecasts.
- Complete Exercises 11, 12, 13, 15 (d,e), 16 (d,e) from Section 9.11 of the book.