SAVE Subcommand (MIXED command)

Use the SAVE subcommand to save one or more casewise statistics to the active dataset.

  • Specify one or more temporary variables, each followed by an optional new name in parentheses.
  • If new names are not specified, default names are generated.

FIXPRED. Fixed predicted values. The regression means without the random effects.

PRED. Predicted values. The model fitted value.

RESID. Residuals. The data value minus the predicted value.

SEFIXP. Standard error of fixed predicted values. These are the standard error estimates for the fixed effects predicted values obtained by the keyword FIXPRED.

SEPRED. Standard error of predicted values. These are the standard error estimates for the overall predicted values obtained by the keyword PRED.

DFFIXP. Degrees of freedom of fixed predicted values. These are the Satterthwaite degrees of freedom for the fixed effects predicted values obtained by the keyword FIXPRED.

DFPRED. Degrees of freedom of predicted values. These are the Satterthwaite degrees of freedom for the fixed effects predicted values obtained by the keyword PRED.

Example

MIXED  SCORE BY SCHOOL CLASS WITH AGE
  /FIXED = AGE
  /RANDOM = SCHOOL CLASS(SCHOOL)
  /SAVE = FIXPRED(BLUE) PRED(BLUP) SEFIXP(SEBLUE) SEPRED(SEBLUP).
  • The SAVE subcommand appends four variables to the active dataset: BLUE, containing the fixed predicted values, BLUP, containing the predicted values, SEBLUE, containing the standard error of BLUE, and SEBLUP, containing the standard error of BLUP.