Prev: seed setting
Next: modeling time series data using Generalized Linear Model in SAS
From: hansi_m on 8 Mar 2010 16:57
Hi,

I was wondering if it is possible to name the name for the score under
PROC SCORE. Usually I get the same name as the target variable with an
extension (for example model1, model2...). In the following case I
even got no name for the score (quite a surprize for me). Here it is
the log:

18 proc score data=newboot1 score= lifereg_est
out=lifereg_scored type=parms;
19 var x-x10;
22 run;

WARNING: Both _NAME_ and _MODEL_ variables are found in the SCORE=
data set. The _MODEL_ variable will be used to name the score
variables.
WARNING: The name for score 1 is blank.

Any thoughts,

Thanks,
Hansi
From: Robin R High on 9 Mar 2010 14:38
Hansi,

You can either drop the _model_ variable when you call PROC LIFEREG:

proc lifereg data=indat outest=OUTEST(drop=_model_);

or

rename the _model_ variable when you call PROC SCORE:

proc score data=newboot1 score= lifereg_est(rename=(_model_=mdl)) etc..

The scored variable from PROC SCORE will then have the name given in the
_name_ variable of the outest dataset.


Robin High
UNMC






From:
hansi_m <mytkolli(a)GMAIL.COM>
To:
SAS-L(a)LISTSERV.UGA.EDU
Date:
03/09/2010 10:13 AM
Subject:
proc score
Sent by:
"SAS(r) Discussion" <SAS-L(a)LISTSERV.UGA.EDU>



Hi,

I was wondering if it is possible to name the name for the score under
PROC SCORE. Usually I get the same name as the target variable with an
extension (for example model1, model2...). In the following case I
even got no name for the score (quite a surprize for me). Here it is
the log:

18 proc score data=newboot1 score= lifereg_est
out=lifereg_scored type=parms;
19 var x-x10;
22 run;

WARNING: Both _NAME_ and _MODEL_ variables are found in the SCORE=
data set. The _MODEL_ variable will be used to name the score
variables.
WARNING: The name for score 1 is blank.

Any thoughts,

Thanks,
Hansi
 | 
Pages: 1
Prev: seed setting
Next: modeling time series data using Generalized Linear Model in SAS