FORECAST.ETS.STAT function

Returns the statistics behind an ETS forecast (smoothing parameters and error measures) so you can judge how good it is.

Works in desktop Excel only

Syntax

=FORECAST.ETS.STAT(values, timeline, statistic_type, [seasonality], [data_completion], [aggregation])

How to use FORECAST.ETS.STAT in Excel, step by step

FORECAST.ETS.STAT doesn’t run in a web practice sheet because it uses Excel’s own exponential smoothing engine (AAA version of ETS). Follow these steps in Excel itself.

You’ll need

Excel 2016 or later
An evenly spaced timeline

Lay out the history. Dates in A2:A37, values in B2:B37.
Choose a statistic. 1 Alpha, 2 Beta, 3 Gamma, 4 MASE, 5 SMAPE, 6 MAE, 7 RMSE, 8 step size.
Write the formula. For the average error type =FORECAST.ETS.STAT($B$2:$B$37,$A$2:$A$37,6).
Compare models. Lower MAE/RMSE means a closer fit. Try different seasonality values and compare.

Example

=FORECAST.ETS.STAT($B$2:$B$37,$A$2:$A$37,7)

Result: RMSE, e.g. 8.42

MASE below 1 means the model beats a simple “same as last time” forecast.