Tutorial learning objectives - Week 5

By the end of this tutorial, you should be able to:

1. Describe the forecasting workflow and understand the sequence of steps involved in building and evaluating forecasting models.

2. Apply benchmark forecasting methods including mean, naïve, seasonal naïve (snaive), and drift methods and understand their appropriate use cases.

3. Interpret residual diagnostics using visual tools such as residual plots and histograms to assess forecast accuracy and model performance. Understand fitted values and residuals.

4. Understand how forecast uncertainty is represented through distributions and how this connects to histograms of residuals.

5. Impact of transformations on the distribution shape, mean, and median.

6. Combine decomposition and forecasting.

7. Evaluate point forecast accuracy inlcuding understanding forecast error metrics such as RMSE, MAPE, MASE, RMSSE.