Select variables dataset

[Thalitacosta]

I have many variables in a database by date: Name variables: XXXX_20220101 YYYY_20220101 ZZZZ_20220101 WWWW_20220101 XXXX_20220202 YYYY_20220202 ZZZZ_20220202 WWWW_20220202 XXXX_20220303 YYYY_20220303 ZZZZ_20220303 WWWW_20220303 ... I want to select all the variables of a date in a new database. How do I do it? where scan(all,-1,'_') = 20230612;

[Tom]

Explain more about how the data is structured.  Why do you have dates stored in the NAME of the variable.  It is best to store data in the VALUE of a variable instead.  

If you have a variable (or set of variables) that uniquely identify the observations in your dataset perhaps you should first transpose the data.

proc transpose data=have out=tall;
  by id;
run;

Then you can select the observations from the TALL dataset what came from variables that had a particular pattern to their names.

data want;
  set tall;
  where _name_ like '%^_20230612' escape '^' ;
run;

If you really want to perpetuate the problem of having dates in variable names you could build a list of variable names into a macro variable

proc sql noprint;
select nliteral(name) into :varlist separated by ' '
 from dictionary.columns
 where libname='WORK' and memname='HAVE'
   and name like '%^_20230612' escape '^'
;
quit;

and use the macro variable to generate part of the code to select the variables of interest. 

data want;
  set have;
  keep id &varlist;
run;