FORECAST.ETS.STAT

This function return a statistical value as a result of time series forecasting. Statistic type indicates which statistic is requested by this function.

Syntax

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

Arguments

The function has the following arguments:

Argument

Description

values

[Required] is a range of the historical values for which you want to predict a new point.

timeline

[Required] is a range of date/time values that correspond to the historical values. The timeline range must be of the same size as the values range. Date/time values must have a constant step between them.

statistic_type

[Required] A numeric value between 1 and 8, indicating which statistic will be returned for the calculated forecast.

The table below shows the eight possible statistical values and their corresponding results.

Value	Statistic type	Description
1	Alpha	It is the base parameter of ETS algorithm. Higher values indicate more weight to recent data.
2	Beta	It is the trend parameter of ETS algorithm. Higher values indicate more weight to recent trends.
3	Gamma	It is the seasonality parameter of ETS algorithm. Higher values indicate more weight to recent seasonal periods.
4	MASE	It stands for Mean Absolute Scaled Error metric, a measure of forecast accuracy.
5	SMAPE	It stands for Symmetric Mean Absolute Percentage Error metric, an accuracy measure based on percentage errors.
6	MAE	It stands for the Mean Absolute percentage Error metric, a measure of accuracy based on percentage errors.
7	RMSE	It stands for Root Mean Squared Error metric, a measure of differences between predicted and observed values.
8	Step Size	It is detected in the historical data timeline.

seasonality

[Optional] is a numeric value that specifies which method should be used to detect the seasonality. The possible values are listed in the table below.

1 or omitted: Seasonality is detected automatically. Positive, whole numbers are used for the length of the seasonal pattern.
0: No seasonality, the prediction will be linear.
an integer greater than or equal to 2: The specified number is used for the length of the seasonal pattern.

data_completion

[Optional] is a numeric value that specifies how to process the missing data points in the timeline data range. The possible values are listed below.

1 or omitted: Missing points are calculated as the average of the neighboring points.
0: Missing points are treated as zero values.

aggregation

[Optional] is a numeric value that specifies which function should be used to aggregate identical time values in the timeline data range. The possible values are listed below.

Numeric Value	Function
1 or omitted	AVERAGE
2	COUNT
3	COUNTA
4	MAX
5	MEDIAN
6	MIN
7	SUM

Syntax

Arguments

Data Types

Examples

Version Available