Statistics of Fit (OUTSTAT)

This table contains the statistics for the selected model of each time seriesan aggregation of transactional data into specified time intervals and sorted according to unique combinations of the default attributes (BY variables). This table includes all of the BY variables assigned to the project, variables that identify the selected model, the region over which the fit statistics are computed, and variables for the statistics of fit. The statistics of fit can be used for model selection and are described in Descriptions of Model Selection Criteria. For more information, see the description of the OUTSTAT Object.

Note: For hierarchical forecasting, the statistics of fit in this table show the measures after reconciliation.

The statistics in this table are listed on the Data tab and in the Time Series Viewer and Forecast Viewer as Forecast attributes. You can use these attributes to create 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. in all of the viewers. See Creating Filters for more information.

There are two forecasta numerical prediction of a future value for a specified time period for each unique combination of BY variable values attributes that are not included in the OUTSTAT table but are computed separately. These are described below.

Forecast Attributes That Are Not Part of the OUTSTAT Table
Label	Variable Name	Description
Flat Forecast	_FLAT	Indicates the presence of flat forecasts. 1 indicates that the forecast for the time series has a range close to 0 0 indicates that the forecast for the time series does not have a range close to 0 Forecasts for a time series are determined to be flat when the forecast range (maximum - minimum predicted value) is less than or equal to 0.00001. The _FLAT attribute differs from the flat forecasts in the Execution Summary for a modeling node. The _FLAT attribute includes all time series where the value could be positive, negative, or zero.
Almost Flat Forecast	_ALMOST_FLAT	indicates the presence of almost flat forecasts, as defined on the Reporting tab in the Project Settings. 1 indicates that the forecast for the time series falls within the boundaries of the almost flat definition 0 indicates that the forecast for the time series does not fall within the boundaries of the almost flat definition

The following table shows the variables in the OUTSTAT table, except for the statistics of fit.

Statistics of Fit
Variable Name	Variable Label	Description
_NAME_	Variable Name	the name of the dependent variable
NOBS	Number of Observations	the number of the qualifying subset of observations in the specified model region (either FIT region or FORECAST region) that are potentially available for the computation of fit statistics
N	Number of Observations Used	the actual number of nonmissing observations from the qualifying subset of observations in the specified model region that were actually used to compute fit statistics
NMISSA	Number of Missing Actual Values
NMISSP	Number of Missing Predicted Values
NPARMS	Number of Model Parameters
_REGION_	Region	specifies the time region in which the statistics were collected. Select from the following. FIT indicates that fit statistics were calculated over the fit region FORECAST indicates that fit statistics were calculated over the forecast region If an out-of-sample region is specified, it is included in the forecast region. See Using an Out-of-Sample Region for more information.
_SELECT_	Selection List	the name of model selection list to which the specified _MODEL_ belongs
_MODEL_	Model	the name of the model specification
DFE	Degrees of Freedom Error

Last updated: March 16, 2026