DOCUMENTATION
Sign In

Get access to My SAS, trials, communities and more.

SAS Forecast Server Procedures: User's Guide
 

The HPFENGINE Procedure

STOCHASTIC Statement

  • STOCHASTIC variable-list / options;

The STOCHASTIC statement lists the numeric variables in the DATA= data set whose accumulated values are used as stochastic input in the forecasting process.

Future values of stochastic inputs do not need to be provided. By default, they are automatically forecast using one of the following smoothing models:

  • simple

  • double

  • linear

  • damped trend

  • seasonal

  • Winters method (additive and multiplicative)

The model with the smallest in-sample MAPE is used to forecast the future values of the stochastic input. Information on the model selected for the independent variables can be found in the OUTEST data set.

A data set variable can be specified in only one STOCHASTIC statement. Any number of STOCHASTIC statements can be used.

The following options can be used with the STOCHASTIC statement:

ACCUMULATE=option

specifies how the data set observations are accumulated within each time period for the variables listed in the STOCHASTIC statement. If the ACCUMULATE= option is not specified in the STOCHASTIC statement, accumulation is determined by the ACCUMULATE= option of the ID statement. See the ID statement ACCUMULATE= option for more details.

SELECTION=option

specifies the selection list used to forecast the stochastic variables. The default is BEST, found in SASHELP.HPFDFLT.

REQUIRED=YES | NO

enables or disables a check of inputs to models. The kinds of problems checked include the following:

  • errors in functional transformation

  • an input consisting of only a constant value or all missing values

  • errors introduced by differencing

  • multicollinearity among inputs

If REQUIRED=YES, these checks are not performed and no inputs are dropped from a model. The model might subsequently fail to fit during parameter estimation or forecasting for any of the reasons in the preceding list.

If REQUIRED=NO, inputs are checked and those with errors, or those judged collinear, are dropped from the model for the current series and task only. No changes are kept in the model specification.

This option has no effect on models with no inputs.

The default is REQUIRED=YES.

REPLACEMISSING

specifies that embedded missing actual values be replaced with one-step-ahead forecasts in the STOCHASTIC variables.

SETMISSING=option | number

specifies how missing values (either actual or accumulated) are assigned in the accumulated time series for variables listed in the STOCHASTIC statement. If the SETMISSING= option is not specified in the STOCHASTIC statement, missing values are set based on the SETMISSING= option of the ID statement. See the ID statement SETMISSING= option for more details.

TRIMMISS=option

specifies how missing values (either actual or accumulated) are trimmed from the accumulated time series for variables listed in the STOCHASTIC statement.

If the TRIMMISS= option is not specified in the STOCHASTIC statement, missing values are set based on the TRIMMISS= option of the ID statement. See the ID statement TRIMMISS= option for more details.

ZEROMISS=option

specifies how beginning and/or ending zero values (either actual or accumulated) are interpreted in the accumulated time series for variables listed in the FORECAST statement. If the ZEROMISS= option is not specified in the STOCHASTIC statement, missing values are set based on the ZEROMISS= option of the ID statement. See the ID statement ZEROMISS= option for more details.

Last updated: March 05, 2026