The REGSELECT Procedure
MODEL Statement
MODEL dependent=<effects> / <model-options>;
The MODEL statement names the dependent variable and the explanatory effects, including covariates, main effects, interactions, and nested effects. If you omit the explanatory effects, the procedure fits an intercept-only model.
After the keyword MODEL, the dependent (response) variable is specified, followed by an equal sign. The explanatory effects follow the equal sign.
For information about constructing the model effects, see the section Specification and Parameterization of Model Effects in Chapter 2, Shared Concepts.
The model-options control other aspects of model formation and inference. Table 5 summarizes these options.
Table 5: MODEL Statement Options
| Option | Description |
|---|---|
| Model Options | |
| CLB | Requests confidence limits |
| INCLUDE= | Includes effects in all models for model selection |
| INFORMATIVE | Models missing values by using extra indicator variables |
| NOINT | Suppresses the intercept |
| SS3 | Displays the Type 3 or joint tests of effects |
| START= | Includes effects in the initial model for model selection |
| STB | Displays standardized estimates |
| TOL | Displays tolerance values for the estimates |
| VIF | Displays variance inflation factors for the estimates |
| XPX | Displays sums-of-squares and crossproducts matrix |
Model Options
You can specify the following options in the MODEL statement after a slash (/):
- CLB
requests the
% upper and lower confidence limits for the parameter estimates. By default, the 95% limits are computed; you can use the ALPHA= option in the PROC REGSELECT statement to change the 
level. The CLB option is not available when you specify METHOD=LAR, METHOD=LASSO, METHOD=ELASTICNET, or METHOD=BESTSUBSET in the SELECTION statement. -
INCLUDE=n
INCLUDE=single-effect
INCLUDE=(effects) forces effects to be included in all models. If you specify INCLUDE=n, then the first n effects that are listed in the MODEL statement are included in all models. If you specify INCLUDE=single-effect or INCLUDE=(single-effect), then the specified effects are forced into all models. The INCLUDE= option is not available when you specify METHOD=LAR, METHOD=LASSO, or METHOD=ELASTICNET in the SELECTION statement. For METHOD=BESTSUBSET, only the first n effects that are listed in the MODEL statement are allowed.
- INFORMATIVE
models missing values by using extra model effects. These effects consist of dummy variables that take the value 1 when the value of a continuous model variable involved in the effect is missing, and take the value 0 otherwise. The missing value in the original model effect is replaced by the average value of the effect for the nonmissing values. For continuous-by-class effects, such as A*x, where A is a classification variable and x is a continuous variable, informative missingness creates multiple dummy columns and substitutes the effect mean of x that corresponds to the respective level of A. Missing values for classification variables are treated as valid levels. For more information about informative missingness, see the section Informative Missingness in Chapter 2, Shared Concepts.
- NOINT
suppresses the intercept term that is otherwise included in the model.
- SS3
displays a "Model Anova" table that contains tests computed with Type III sums of squares for each effect. For more information, see the section Joint Tests and Type 3 Tests. This option is not available when you specify METHOD=BESTSUBSET in the SELECTION statement.
-
START=n
START=single-effect
START=(effects) is used to begin the selection process in the FORWARD, FORWARDSWAP, and STEPWISE selection methods from the initial model that you designate. If you specify START=n, then the starting model consists of the first n effects that are listed in the MODEL statement. If you specify START=single-effect or START=(single-effect), then the starting model consists of these specified effects. The START= option is not available when you specify METHOD=BACKWARD, METHOD=LAR, METHOD=LASSO, METHOD=ELASTICNET, or METHOD=BESTSUBSET in the SELECTION statement.
- STB
produces standardized regression coefficients. A standardized regression coefficient is computed by dividing a parameter estimate by the ratio of the sample standard deviation of the dependent variable to the sample standard deviation of the regressor. This option is not available when you specify METHOD=BESTSUBSET in the SELECTION statement.
- TOL
produces tolerance values for the estimates. Tolerance for a parameter is defined as
, where 
is obtained from the regression of the parameter on all other parameters in the model. This option is not available when you specify METHOD=LAR, METHOD=LASSO, METHOD=ELASTICNET, or METHOD=BESTSUBSET in the SELECTION statement. - VIF
produces variance inflation factors along with the parameter estimates. Variance inflation is the reciprocal of tolerance. This option is not available when you specify METHOD=LAR, METHOD=LASSO, METHOD=ELASTICNET, or METHOD=BESTSUBSET in the SELECTION statement.
- XPX
displays the
crossproducts matrix for the model. The crossproducts matrix is bordered by the 
and 
matrices.