To construct a purely seasonal model for the following monthly series, we need to identify binary variables that capture the relevant effects:
a. US de-trended monthly Salmonella case numbers exhibit uniformity across each month within a year. One binary variable could be assigned to each month, indicating whether it belongs to a specific month of the year or not.
b. For the campus bookstore's sales, a binary variable can represent each month within a semester. For instance, if we're considering only the first, second, third, and fourth months of the spring and fall semesters, we would assign binary values to each month indicating whether it falls within those specified months or not.