CRITERIA Subcommand (LINEAR_LASSO extension command)

CRITERIA allows specification of several parameters that control analyses.

Intercept = True | False.

Whether or not to include an intercept in the fitted models. Note that the extension procedure does not center or standardize the dependent variable, and the intercept is not penalized during estimation. The intercept can be penalized by specifying False and including a constant predictor value (this is generally not recommended). The default value is True.

Standardize = True | False.

Whether to standardize independent variables. All independent variables are standardized by default. Categorical factors are represented by an indicator or dummy variable for each observed category, and each of these is standardized individually. The default value is True.

TIMER = value.

Number of minutes allowed for model computations to execute. The value must be an integer greater than or equal to 0. When 0 is specified, the timer is turned off. The default value is 5.

NFOLDS = value.

Number of splits or folds for crossvalidation evaluation of models. The value must be an integer value greater than 1. The default value is 5.

STATE = value.

Value of the random_state setting in Python that is used when performing crossvalidation evaluation of models. The setting allows replication of results involving pseudo-random numbers. The value must be an integer in the range of 0 to 2³²-1. The default value is 0.

TRACETABLE = value. (syntax-only)

The maximum number of alpha values that are used to produce a trace data table (used in plots when MODE=TRACE). The setting produces an optional table in trace mode that displays R2, MSE, and regression coefficient values for each specified alpha value. The specified value must be a non-negative integer. The default value is 0 (no trace results). Data for the entire trace displays when the specified value is greater than or equal to the number of alpha values. Data for the specified value displays when the value is less than the number of alpha values.