Dependent Variable

If only a single candidate dependent variable is detected in your data source, it is automatically selected. If multiple candidates are detected, you must choose one.

Dependent variable

Specify the variable in your data set that you want to forecasta numerical prediction of a future value for a specified time period for each unique combination of BY variable values. This variable must be numeric.

Hierarchy aggregation

Aggregation is the process of combining data from more than one time seriesan aggregation of transactional data into specified time intervals and sorted according to unique combinations of the default attributes (BY variables) to form a single series. Select the aggregationthe process of combining more than one time series to form a single series within the same time interval. For example, data can be combined into a total or an average. method that you want to use for all of the time series in each level of the hierarchy. The dependent variable has fewer aggregation options than independent variables. The following examples explain when you might want to use an aggregation method:

  • Your data set contains the sales for a group of products and you want to know the total sales for a category.
  • Your data contains the price of each product and you want to know the average price for a product line.

Select from one of the following options.

Sum of Values

aggregates the vector values based on the number of values.

eh equals q. Click image for alternative formats.
Average of Values

aggregates the vector values based on the average of their values.

eh equals , r with macron above , equals , fraction 1 , over q sub n end fraction . cap sigma with q equals 1 below and with q above . r sub q. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , equals 0. Click image for alternative formats., then a is set to missing.

Note: Average of values applies only to hierarchical modeling. The plot and table in Overrides shows a dynamic sum of values of the time series captured by the selected 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. .
Time interval accumulation

Accumulation combines data within the same time interval into a summary value for that time interval.

Average of Values

accumulates the vector values based on the average of their values.

eh equals , r with macron above , equals , fraction 1 , over q sub n end fraction . cap sigma with q equals 1 below and with q above . r sub q. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , equals 0. Click image for alternative formats., then a is set to missing.

Corrected Sum of Squares

accumulates the vector values based on their corrected sum of squares.

eh equals . cap sigma with q equals 1 below and with q above . open , r sub q , negative , r with macron above , close squared. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , equals 0. Click image for alternative formats., then a is set to missing.

Maximum of Values

accumulates the vector values based on the maximum of their values.

eh equals mehx of open . the set , r sub q end set with subscript q equals 1 , and with superscript q , end sub-superscript . close. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , equals 0. Click image for alternative formats., then a is set to missing.

Minimum of Values

accumulates the vector values based on the minimum of their values.

eh equals min of open . the set , r sub q end set with subscript q equals 1 , and with superscript q , end sub-superscript . close. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , equals 0. Click image for alternative formats., then a is set to missing.

Number of non-missing values

accumulates the vector values based on the number of nonmissing values.

eh equals , q sub n. Click image for alternative formats.
Number of Missing values

accumulates the vector values based on the number of missing values.

eh equals . q sub n m i s s end sub. Click image for alternative formats.
Standard Deviation of Values

accumulates the vector values based on their standard deviation.

eh equals . square root of fraction 1 , over q sub n , minus 1 end fraction . cap sigma with q equals 1 below and with q above . open , r sub q , minus , r with macron above , close squared end root. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , less than or equal to 1. Click image for alternative formats., then a is set to missing.

Sum of Values

accumulates the vector values based on the summation of their values.

eh equals . cap sigma with q equals 1 below and with q above . r sub q. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , equals 0. Click image for alternative formats., then a is set to missing.

Uncorrected Sum of Squares

accumulates the vector values based on their uncorrected sum of squares.

eh equals . cap sigma with q equals 1 below and with q above . open , r sub q , close squared. Click image for alternative formats.

Missing values are ignored in the summation. If q sub n , equals 0. Click image for alternative formats., then a is set to missing.

For more information, see Time Interval Accumulation Settings.

Missing interpretation

Once the data has been accumulated to form a time series, missing value interpretation is performed. If any time series contain missing values for the dependent or independent variables, you can specify how SAS Visual Forecasting should interpret these missing values. For more information, see Interpretation Step.

Choose from the following options:

First

Specifies that missing values are set to the first nonmissing value of all accumulated nonmissing values in the span of the series.

Last

Specifies that missing values are set to the last nonmissing value of all accumulated nonmissing values in the span of the series.

Maximum

Specifies that missing values are set to the maximum value of all accumulated nonmissing values in the span of the series

Median

Specifies that missing values are set to the median value of all accumulated nonmissing values in the span of the series

Minimum

Specifies that missing values are set to the minimum value of all accumulated nonmissing values in the span of the series

Average

Specifies that missing values are set to the average value of all accumulated nonmissing values in the span of the series.

Missing

Specifies missing values to remain missing. Use this option if a missing value indicates an unknown value.

Next

Specifies that missing values are set to the next period’s accumulated nonmissing value. Missing values at the end of the accumulated series remain missing.

Previous

Specifies that missing values are set to the previous period’s accumulated nonmissing value. Missing values at the beginning of the accumulated series remain missing.

0

Specifies missing values to be set to zero. This setting is often used for transactional datatimestamped data collected over time at no particular frequency. Some examples of transactional data are point-of-sale data, inventory data, call center data, and trading data., because no recorded data usually implies no activity.

Last updated: March 16, 2026