Try this: data class;   set sashelp.class;   if (13 <= age <= 15) and sex='F' then delete; run; ... View more

Hi Patrick, You are right. The instructuion I meant was a MERGE instead a SET. I already Edited the original message. My solution gives only four observations and asumes that both datasets are sorted by proc1 proc2. My test program follows: DATA TABLE1; input PROC1 PROC2 $ EDIT; Datalines; 100 G0427 0 102 0311T 0 102 36010 0 102 76998 0 ; run; DATA TABLE2 ;       input CLAIM PROC1 PROC2 $ NET1 NET2 ADJD1 ADJD2 ADJDT1 $10. ADJDT2 $10.; Datalines; 111 100 3032 5 5 111 873 10/10/2013 10/10/2013 222 100 G0427 5 5 111 873 10/11/2013 10/11/2013 333 102 0 19 15 111 873 10/12/2013 10/12/2013 444 102 36010 20 20 111 873 10/13/2013 10/13/2013 ; run; Data need;   merge table2 (in=t2) table1(drop=edit in=t1);   by proc1 proc2;   flag=t1 and t2;   if t2; run; Regards, ... View more

yChanged the SET Try this: data cci.phyclms11;   merge cci.phyclms10 (in=T1) cci.cci0 (drop=edit in=T2);   by proc1 proc2;   flag= t1 and t2;   if t1; run; Regards, ... View more

I understand that SAS/ACCESS Interface for Microsoft SQL Server is only available for UNIX and LINUX Operating Systems. To access SQL Server in Windows Operating System you have to use SAS/ACCES to ODBC or OLE_DB. ... View more

I Think that if you are a SAS Programmer and already have a big number of SAS Programs already running and a good experience in using the DM then the way you use DI Studio makes sense but using DI Studio to develop new programs bring the at least the following benefits: You get the proccess well documented in diagrams. You can use the Impact and Reverse impact Analysis DI is intuitive (very easy) and drag-and-drop for developping ETL processes (in fact its former name was ETL Studio) You can use Change Management techniques within the tool. In summary I think a good balance is using User Written Code transformations ONLY in cases where there is not a transformation for the functionality you need to develop and connect them to the rest of transformation via Tables/Views. ... View more

Shelby, I think I missunderstood the problem so maybe PATHNAME is not the solution. If I understand, you are able to see the .csv file from the EG session. In this case what you have to do is to findout the place (starting path) from where EG is able to view these files using the following procedure in SAS Management Console: Expand: Server Manager --> SASMain --> SASMain - Workspace Server (Properties) --> Options tab --> Advanced Options...  --> File Navigation. In this point you should see three options SAS System Root, System Root or Path. The one that is selected is the left missing part of your path. Good luck. ... View more

You have tried to use the PATHNAME Function?  See: SAS(R) 9.3 Functions and CALL Routines: Reference ... View more

You can simplify the select statement eliminating the do-end blocks innecesary in this case: data when ; set raw ; select (Region); when('Central') code=1 ; when('East') code=2 ; when('West') code=3 ; otherwise code=0 ; end ; run ; ... View more