Filter for BLUP Coefficients

The Filter for BLUP Coefficients field is an optional parameter that enables you to specify a logical expression for subsetting predictors based on their estimated Best Linear Unbiased Prediction (BLUP) coefficients, which are shrunken regression coefficients.

This option is useful when you want to reduce the ridge regression model to include predictors with BLUP coefficients that satisfy a certain constraint. The model is refitted using the reduced set of predictors to produce final results. Note then the newly computed BLUPs might not satisfy the specified constraint.

The expression must be a valid SAS WHERE clause, and it must contain the variable name "blup". For example, to keep predictors with nonnegative BLUPs only, you could use the following WHERE expression:

WHERE blup > 0

Note: Note: the word where has already been entered for you.

For more information about using filter fields, see The SAS WHERE Expression.