Data Attributes Options

Select from these options for data attributes and seasonality:

• Override prediction range: When this option is selected, the prediction overrides the minimum number of historical data points typically required to run a prediction. Instead, it predicts the number of periods based on the number of historical data points that are available. For example, it predicts y periods when 2y historical data points are available.

  If this option is not selected, if the prediction range is set to x periods, a minimum of 2x historical data points are required to run a prediction. If the minimum number of historical data points are not available, the prediction does not proceed.

• Auto Detect: When this option is selected, the prediction automatically detects seasonality based on the data.
• Yearly Seasonal Override (12 periods): When this option is selected, seasonality is defined by the number of level 0 periods from the selected hierarchy. In cases where the number of historical data points is less than 2x of yearly seasonality, the prediction reverts to using Auto Detect for seasonality.
• Custom: When this option is selected, you can override the seasonality with a custom value by providing it in Periods per cycle. You can select this option when seasonality is not yearly.
• Periods per cycle: When Custom is selected, select the exact number of periods to be considered. If not enough data points are available, the prediction reverts to using Auto Detect for seasonality. The maximum allowed value for Periods per cycle is half of the total number of historical data range periods.