1. Create a tsibble with total Holiday travellers for Victoria and Queensland from the tourism data set using the code below. Plot the series. What do you see?

holidays <- tourism |>
  filter(State %in% c("Victoria", "Queensland")) |>
  filter(Purpose == "Holiday") |>
  as_tibble() |>
  summarise(Trips = sum(Trips), .by = c("State", "Quarter")) |> 
  as_tsibble(index = Quarter, key = State)

2. Generate 4 year ahead forecasts from all four benchmarks. Plot them using autoplot(). Comment in the resulting forecasts.

3. Plot the residuals from the most appropriate benchmark using gg_tsresiduals(). What do you see?

4. Test if the residuals are white noise. What do you conclude?

5. Plot point and interval forecasts from the most appropriate benchmark.

6. Now try a decomposition forecasting model.

7. Use accuracy() to evaluate which benchmark fits the data best.

8. Use a test set of last 3 years to check forecast accuracy.

9. Now use time series cross-validation to check forecast accuracy.

10. Exam 2024

    • Section A: Q3, Q6
    • Section B: Q3a, Q3b