Hi guys,          I have a dataset that I got from joining two tables. Since there are multiple users on a given account, I have duplicate values. I am trying to find a way to code the out put listed below. As you see, I want to concatenate records based on a individual employee id and trans_no, acct_no, acct_name, trans_date,trans_amt along with employee name and his dob. I tried below code, but it's not working out as I want to concatenate duplicate part of the record to non-duplicated record (employee fname,lname,dob). data have; input fname $ lname $ trans_no $ emp_id $ acct_no $ acct_name $ trans_date date9. trans_amt dob date9. ; FORMAT DOB TRANS_DATE DATE9.; cards; John Smith 123XER ABCD 12345 JonasLLC 17May2020 1500 02mar1990 Sarah Frost 123XER ABCD 12345 JonasLLC 17May2020 1500 22Jun1995 Noel White 458ABC ABCD 45689 Jonas 11May2020 20000 12may2000 Tim Kerry 458ABC ABCD 45689 Jonas 11May2020 20000 16jun2005 Tom Baker 784TYU ABCD 54321 JonasLLC 17Mar2020 15000 02mar2010 Nick Black 342VCF WXYZ 67890 Chipotle 15May2020 10000 15Jun1978 Mike Meyer 987RTV WXYZ 568471 Macys 18jan2020 45000 14Feb1992 Ronney Black 342VCF WXYZ 67890 Chipotle 15May2020 10000 15Jun2000 Nancy Meyer 987RTV WXYZ 568471 Macys 18jan2020 45000 14Feb2012 ; run; PROC SORT DATA=HAVE ; BY emp_id ; RUN; data want (keep=total); LENGTH TOTAL $500; set have; by emp_id; IF FIRST.EMP_ID THEN TOTAL= STRIP(TOTAL)||'|'||STRIP(emp_id)||'|'||STRIP(trans_no)||'|'||STRIP(acct_no)||'|'||STRIP(acct_name)||'|'||STRIP(PUT(trans_date, date9.))||'|'|| STRIP(PUT(trans_amt,8.))||'|'||STRIP(fname)||'|'||STRIP(lname)||'|'||STRIP(PUT(dob,DATE9.))||'|'||'***'||'|' ; ELSE DELETE; RUN; I want the output to look like this: |ABCD|123XER|12345|JonasLLC|17May2020|1500|***|John|Smith|02mar1990|***|Sarah|Frost|22Jun1995|*** |458ABC|45689|Jonas|11May2020|20000|***|Noel|White|12may2000|***|Tim|Kerry|16jun2005|*** |784TYU|54321|JonasLLC|17Mar2020|15000|Tom|Baker|02mar2010| |WXYZ|342VCF|67890|Chipotle|15May2020|10000|***|Nick|Black|15Jun1978|***|Ronney|Black|15Jun2000|*** |987RTV|568471|Macys|18jan2020|45000|***|Mike|Meyer|14Feb1992|***|Nancy|Meyer|14Feb2012|   Thanks in advance. ... View more

Guys, I have a dataset i have and I want the output to look like this. I want the output to look like this  BBB B2546 C1254 BBB B4578 C12456 BBB B7995 C14576 BBB B1245 C1259 I want to build a logic where on my ID group if there is a match of account1 and account2 then I want that ID group to be deleted (eg: AAA A2345 A2345 & CCC D4568 D4568) and I want to have the ID group that has all different account1 as well as account2 . I tried the code below which didn't work. I appreciate any help on this in advance.    data check; input id $ account1 $ account2 $ ; cards; AAA A1234 B1456 AAA A1234 B74156 AAA A2345 A2345 BBB B2546 C1254 BBB B4578 C12456 BBB B7995 C14576 BBB B1245 C1259 CCC D4568 F1254 CCC D4568 G1458 CCC D4568 D4568 ; RUN; PROC SQL; CREATE TABLE NEW AS SELECT * FROM CHECK GROUP BY ID HAVING ACCCOUNT1 NOT IN (SELECT DISTINCT ACCOUNT2 FROM CHECK ); QUIT; ... View more

Dear SAS Community,           I have macro variables generated from &var1 to &var50. I want to query them in PROC SQL and list them in where clause (like &var1 or &var2 .....or &var50). How can I generate a macro to list all of these 50 variables with 'OR' separation to list. I want to have a single macro to list all of these 50 variables like (%let new= &var1 or &var2 .... &50).  In data step we have : operator to list (like var: will get all of the variables whose prefix changes).  example code:  proc sql; create table new_table as  select * from want  where number in (&new); quit;   thanks     ... View more