CNTSELECT Procedure

SELECTION Statement

SELECTION <METHOD=method <(method-options)>><options>;

The SELECTION statement performs model selection by examining whether effects should be added to or removed from the model according to rules that are defined by the selection methods. The statement is fully documented in the section SELECTION Statement in Chapter 4, Shared Concepts. The CNTSELECT procedure supports only a subset of the method values and method-options that are listed in that section.

You cannot specify both the SELECTION statement and the BAYES statement.

You can use the following option to specify the selection method:

METHOD=BACKWARD | FORWARD | FORWARDSWAP | STEPWISE

specifies the selection method to use. You can specify one of the following methods:

BACKWARD: specifies the backward elimination method.
FORWARD: specifies the forward selection method.
FORWARDSWAP: specifies forward-swap selection, which is an extension of the forward selection method.
STEPWISE: specifies the stepwise regression method.

By default, METHOD=STEPWISE. For more information about each of the selection methods, see the section Model Selection Methods

PROC CNTSELECT supports a specific set of values for the following method-options:

CHOOSE=criterion

chooses from the list of models (at each step of the selection process) the model that yields the best value of the specified criterion. PROC CNTSELECT supports only the following values for criterion:

AIC: specifies Akaike’s information criterion.
SBC: specifies the Schwarz Bayesian information criterion.

By default, the value of the CHOOSE= option is the same as the value that you assign to the SELECT= option. For more information, see the detailed description of the CHOOSE= option.

SELECT=criterion

specifies the criterion that the procedure uses to determine the order in which effects enter or leave at each step of the selection method. PROC CNTSELECT supports the following values for criterion:

AIC: specifies Akaike’s information criterion.
SBC: specifies the Schwarz Bayesian information criterion.

By default, SELECT=SBC. For more information, see the detailed description of the SELECT= option.

STOP=criterion

specifies a criterion to use to stop the selection process. PROC CNTSELECT supports only the following values for criterion:

AIC: specifies Akaike’s information criterion.
SBC: specifies the Schwarz Bayesian information criterion.

By default, the value of the STOP= option is the same as the value that you assign to the SELECT= option. For more information, see the detailed description of the STOP= option.

PROC CNTSELECT supports the following method-options, which are described in the section SELECTION Statement in Chapter 4, Shared Concepts:

PROC CNTSELECT supports the following statement options, which are described in the section SELECTION Statement in Chapter 4, Shared Concepts:

Last updated: November 24, 2025