BookmarkSubscribeRSS Feed
dsriggs
Fluorite | Level 6

Is there a way I can name the forecasted values from Proc Arima in the out dataset?  For example with Proc Reg I could do something like:

proc reg data=data outest=reg noprint;
name:model y=x;
run;

proc score data=data score=reg type=parms predict out=pred;
var x;
run;

And the forecasted values will be called "name".  Is there a way to do something like this with Proc Arima?  Whenever I use the out statement it stores the forecasts as "forecast of y".

2 REPLIES 2
alexchien
Pyrite | Level 9

the "forecast of y" is actually the lable of the column, not the column name. The standard column name produced by proc arima is FORECAST. You can use the standard sas RENAME statement to change the column name of the output table. For example:

 

/*name the forecast column "xyz"*/

proc arima data=sashelp.air;
    identify var=air(1,12);
    estimate q=(1)(12) noint method=ml;

    forecast id=date interval=month out=b (rename=(forecast = xyz));
run;
quit;

 

udo_sas
SAS Employee

Hello -

In addition to @alexchien's response I'd like to add: as ARIMA creates a common SAS data set one can use utility procedures such as DATASETS to modify attributes such as labels and names. For example if you would like to change both name and label of a variable produced by ARIMA this code snippet might be of interest.

Thanks,

Udo

 

proc arima data=sashelp.air;

identify var=air(1,12);

estimate q=(1)(12) noint method=ml;

forecast id=date interval=month out=work.want;

run;

proc datasets lib=work memtype=data nodetails nolist;

modify want;

attrib forecast label="Hello World";

rename forecast=MYNAME;

quit;

sas-innovate-white.png

Our biggest data and AI event of the year.

Don’t miss the livestream kicking off May 7. It’s free. It’s easy. And it’s the best seat in the house.

Join us virtually with our complimentary SAS Innovate Digital Pass. Watch live or on-demand in multiple languages, with translations available to help you get the most out of every session.

 

Register now!

Discussion stats
  • 2 replies
  • 1547 views
  • 0 likes
  • 3 in conversation