Hello,
I am trying to run multiple SAS forecasts without using TSFS.
Would it be possible to have the SAS code which would replicate the following model (the option bellow are the one I would use in TSFS Develop Models):
(I'm using SAS 9.1.3)
Interval: Month
Period of Fit: Jan2008 to Mar2012
Period of Evaluation: Jan2012 to Mar2012
Forecast Horizon: 44 (to Dec2015)
Error Method: Mean Absolute Percent Error
Smoothing Model: Winters Method Additive
Smoothing Weights:
Level: Optimize
Trend: Optimize
Season: Optimize
Bounds: Zero One/Additive
No Transformation
Thank you,
Madalin
Update: please note that ESM was introduced with SAS 9.2: http://support.sas.com/kb/31/654.html.
Hello -
I would suggest to have a look at PROC ESM.
Example:
* Create a test data set using your specifications;
data work.have(where=(date le "01MAR2012"d));
set sashelp.air;
date=intnx("month",date,59*12);
run;
*run ESM with Additive Winters;
proc esm data=work.have lead=44 outfor=work.want plot=forecasts;
id date interval=month;
forecast air / model=addwinters;
run;
Thanks,
Udo
Update: please note that ESM was introduced with SAS 9.2: http://support.sas.com/kb/31/654.html.
Hello -
I would suggest to have a look at PROC ESM.
Example:
* Create a test data set using your specifications;
data work.have(where=(date le "01MAR2012"d));
set sashelp.air;
date=intnx("month",date,59*12);
run;
*run ESM with Additive Winters;
proc esm data=work.have lead=44 outfor=work.want plot=forecasts;
id date interval=month;
forecast air / model=addwinters;
run;
Thanks,
Udo
Thank you Udo,
Much appreciated.
Proc esm is the one I need but unfortunately I'm using SAS 9.1.3.
Regards,
Madalin
SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!
Learn how to run multiple linear regression models with and without interactions, presented by SAS user Alex Chaplin.
Find more tutorials on the SAS Users YouTube channel.