Predicted Score for Logistic Regression in MODEL HANDLE

[Eric Black]

I am running a logistic regression and want to save the model so I can use
“model handle” afterwards.  My syntax is something like this below:

LOGISTIC REGRESSION VARIABLES var01
           /METHOD=enter
 Var02 var 03
            /save Pred(Pred_Model) PGROUP(PGROUP_Model)
           /OUTFILE= PARAMETER ('model.xml').

MODEL HANDLE NAME=Model1 FILE=model.xml'
/OPTIONS MISSING=SUBSTITUTE.
COMPUTE model =APPLYMODEL(Model1, 'predict').
 exe.
MODEL CLOSE NAME=Model1.

It is my understanding from reading the manual that SAVE PGROUP is the same
as ‘predict’ in MODEL HANDLE.  I am actually also interested in the score
(which is SAVE PRED for LOG REGRESSION), but what would that be for MODEL
HANDLE?  It’s not PROBABILITY from what I understand.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD