13.6 Exercises

  1. Compare STL and Dynamic Harmonic Regression forecasts for one of the series in the pedestrian data set.

    1. Try modifying the order of the Fourier terms to minimize the AICc value.
    2. Check the residuals for each model. Do they capture the available information in the data?
    3. Which of the two sets of forecasts are best? Explain.
  2. Consider the weekly data on US finished motor gasoline products supplied (millions of barrels per day) (series us_gasoline):

    1. Fit a dynamic harmonic regression model to these data. How does it compare to the regression model you fitted in Exercise 5 in Section 8.10?
    2. Check the residuals from both models and comment on what you see.
    3. Could you model these data using any of the other methods we have considered in this book? Explain why/why not.
  3. Experiment with using NNETAR() on your retail data and other data we have considered in previous chapters.