464 Statistical Methods
Additive Seasonality
Sometimes the seasonal variation is expressed in additive terms, especially
if there is not much growth. If the highest annual sales total is no more than
twice the lowest annual sales total, it probably does not matter whether you
use differences or ratios. If you can express the month-to-month changes in
additive terms, the seasonal variation is called additive seasonality. Addi-
tive seasonality is expressed in terms of differences from the expected aver-
age for the year. In Table 11-5, the seasonal adjustment for December sales is
2 240, resulting in an adjusted sales for that month of 681. After adjustment
for the time of the year, December turned out to be one of the most successful
months, at least in terms of exceeding goals.
Table 11-5 Additive Seasonality
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Sales 298 378 373 443 374 660 1004 1153 1388 904 715 441
Factor 2325 2270 2270 22002280 255 350 450 550 220 70 2240
Adjusted
sales
623 648 643 643 654 715 654 703 838 684 645 681
In this chapter you’ll work with multiplicative seasonality only, but you
should be aware of the principles of additive seasonality.
Seasonal Example: Liquor Sales
Are liquor sales seasonal? The Liquor workbook has monthly liquor store
sales in terms of millions of dollars from January 1996 through December
- The workbook contains the variables and reference names shown in
Table 11-6.
Table 11-6 The Liquor Workbook
Range Name Range Description
Year A2:A145 The year
Month B2:B145 The month
Year_Month C2:C145 The year and the month
Sales D2:D145 The monthly liquor sales in millions of dollars