Syntax :

=FORECAST.ETS( target_date, values, timeline, [seasonality], [data completion], [aggregation] )

Parameters :

• The function parameters are as follows:

target date –

• A date/time for which a value should be predicted.
  (Must be after the most recent date/time in the historical timeline.)

values –

• An array of previously known historical values for which you wish to anticipate the future point.

Timeline –

• The independent array of dates/times that relate to each value.
  This must meet the following requirements:
  The timeline array must be the same length as the values array; however, the dates/times in the timeline must have a constant step length between them.                                                                        Depending on the value of the [data completeness] option, up to 30% of points may be missing and dealt with.                                                                                                                                                                      The timeline may contain duplicates, whose associated values will be aggregated according to the [aggregation] option.                                                                                                                                             The timeline’s dates and times can be in any sequence.

[seasonality] –

• An optional input specifying the algorithm to be used to detect seasonality in data.
  If provided, this must be a positive integer between 0 and 8784, corresponding to:

[data completion] –

• An optional input that specifies how the algorithm should handle missing timeline points.
  If provided, the [data completion] parameter can have a value of 0 or 1, which means:

[aggregation] –

• An optional parameter that tells the algorithm how to aggregate data with the same timestamp.
  If provided, this can be any integer between 1 and 7, which means: