SAS Visual Forecasting detects flat forecasts in a project. Flat forecasts are horizontal forecasts with a slope of zero. You can also set a tolerance level to define almost flat forecasts for your project, which are horizontal forecasts with a slope near zero.
Time series are identified as flat forecasts when all forecasta numerical prediction of a future value for a specified time period for each unique combination of BY variable values values fall within a very small predefined range (when the maximum of the forecast values minus the minimum of the forecast values is less than 0.00001). You can also set a tolerance level to define almost flat forecasts for your project, where the forecasts have minimal variation or trend over the forecast horizonthe number of intervals into the future, beyond a base date, for which analyses and predictions are made.. Depending on the method selected, almost flat detection evaluates forecast range, standard deviation, or a combination of both standard deviation and linear trend (slope).
To define almost forecasts for your project, open Project Settings and select the Reporting tab on the left. Use the settings on this tab to define the tolerance level for identifying almost flat forecasts. See Reporting Settings for more information.
To show the flat or almost flat forecast time series in the Forecast Viewer, add Forecast attributes (Others) to the Filters pane, and then select the Flat Forecasts or Almost Flat Forecasts filtersa set of specified criteria that are applied to data in order to identify the subset of data for a subsequent operation, such as continued processing. . Each attribute can have a value of yes (1) or no (0).
This feature is available starting in release
2026.01. If you have a project that is upgraded from a previous release, you should
click
the Run all pipelines icon () in the project toolbar so that these settings take effect.