Works beautifully - I don't think I've ever used Proc summary - and am not very data step savvy, except what I parse together from Ask Google 🙂 ! So thank you. Also thank you for the note and instructions on the Excel and transforming that into code to post! Will certainly keep that in mind! ... View more

Does below return what you're after? data have; infile datalines truncover dsd; input Acct_Num :$10. Statement_Code :$1. (Eff_From_Dt Eff_To_Dt) (:ddmmyy10.); format Eff_From_Dt Eff_To_Dt ddmmyy10.; datalines; 123456,,30/06/2016,27/02/2017 123456,,28/02/2017,4/12/2017 123456,,5/12/2017,8/02/2018 123456,,9/02/2018,26/06/2018 123456,T,27/06/2018,9/07/2018 123456,T,10/07/2018,13/07/2018 123456,T,14/07/2018,26/11/2018 123456,T,27/11/2018,30/12/2018 123456,T,31/12/2018,22/01/2019 123456,,23/01/2019,28/03/2019 123456,,29/03/2019,30/05/2019 123456,,31/05/2019,3/06/2019 123456,,4/06/2019,27/08/2019 123456,,28/08/2019,23/10/2019 123456,,24/10/2019,30/10/2019 123456,T,31/10/2019,10/04/2021 123456,T,11/04/2021,30/08/2021 123456,T,31/08/2021,31/12/9999 ; proc sort data=have; by Acct_Num Eff_From_Dt; run; data want(drop=_:); set have; by Acct_Num Statement_Code notsorted; retain _Eff_From_Dt; if first.Statement_Code then _Eff_From_Dt=Eff_From_Dt; if last.Statement_Code then do; Eff_From_Dt=_Eff_From_Dt; output; end; run;   ... View more

probably @Astounding is the way to go. but in sql you can do this as  proc sql; select region, center , assoc, max(metric1) as metric1, max(Descriptor1) as Descriptor1, max(metric2) as metric2, max(Descriptor2) as Descriptor2, max(metric3) as metric3, max(Descriptor3) as Descriptor3, max(metric4) as metric4, max(Descriptor1) as Descriptor4 from (select coalesce(a.Region,b.region,c.region,d.region) as region , coalesce(a.Center,b.center,c.center,d.center) as center, coalesce(a.Assoc, b.assoc, c.assoc,d.assoc) as assoc , Metric1 , Descriptor1, Metric2 , Descriptor2, Metric3 , Descriptor3, Metric4 , Descriptor4 from table1 a full join table2 b on a.Region=b.region and a.Center=b.center and a.Assoc= b.assoc full join table3 c on a.Region=c.region and a.Center=c.center and a.Assoc= c.assoc full join table4 d on a.Region=d.region and a.Center=d.center and a.Assoc= d.assoc)a group by region, center, assoc ; ... View more

You can get EG to generate a slightly more "reusable" file for you by clicking the Performance button, then "Bypass the data cleansing process" on the Import Data task.   This will tell EG to not "fix" the file for you by changing the delimiter and quoting in the fields.  It will instead copy/import the file as-is, which might include a few data processing warts -- but hey, that's life!   ... View more

You're missing a semicolon at the end of line 19. ... View more

Please note that EG projects can only be scheduled on the computers EG is installed on. These would typically be PCs. If you want to schedule jobs on a SAS server then you need to export your EG projects to SAS programs and then schedule the programs. themselves.   Server-based scheduling is much more reliable and is more easily monitored and can be easily set up via SAS Management Console. There is still a case for PC scheduling if it is mainly for your own work purposes.            ... View more