Problem
Australian Wine Sales. Figure 17.19 shows time plots of monthly sales of six types of Australian wines for 1980-1994. The units are thousands of liters. You are hired to obtain short term forecasts (2-3 months ahead) for each of the six series, and this task will be repeated every month.
a. Which forecasting method would you choose if you had to choose the same method for all series? Why?
b. Fortified wine has the largest market share of the above six types of wine. You are asked to focus on fortified wine sales alone, and produce as accurate as possible forecasts for the next 2 months.
• Start by partitioning the data: use the period until December 1993 as the training set.
• Fit a regression model to sales with a linear trend and additive seasonality.
i. Create the "actual vs. forecast" plot. What can you say about the model fit?
ii. Use the regression model to forecast sales in January and February 1994.
c. Create an ACF plot for the residuals from the above model until lag 12. Examining this plot (only), which of the following statements are reasonable conclusions?
• Decembers (month 12) are not captured well by the model.
• There is a strong correlation between sales on the same calendar month.
• The model does not capture the seasonality well.
• We should try to fit an autoregressive model with lag 12 to the residuals.
