What season length should your forecast use?

Seasonality you can name is seasonality you can forecast. The ACF correlates a series with itself shifted back 1, 2, 3 steps and further, and a spike at lag k is a season of length k signing its name.

Analytics Starter Forecasting free

After you install, this is the model to open.

Does Your Revenue Have a Season?

  1. In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
  2. Click Start from a template and put that name in the search box.
  3. Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.

The answer

The four-beat cycle writes its signature across the lags:

Lag 4
+0.74 the season
Lag 8
+0.50 the echo
Lag 2
-0.84 opposite phase

Positive at the season length and its multiples, negative halfway between: that symmetric pattern is what genuine seasonality looks like, and it hands you the one number Holt-Winters asks for, season length 4. On monthly data look for the spike at 12. And if no lag stands out beyond the confidence band, stop forecasting seasons your data does not have.

The model

Sixteen quarters of revenue with a strong Q1, a weak Q3, and no trend. The ACF is computed out to lag 8, two full years of self-comparison.

History16 quarters, $73K to $134K
Max lag8 quarters
ACF per lagcalculated

Once it is in your sheet

  1. The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
  2. Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
  3. Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.

Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.