Date:         Thu, 15 Apr 2004 17:06:01 +0200
Reply-To:     Ace <b.rogers@VIRGIN.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Ace <b.rogers@VIRGIN.NET>
Subject:      Re: variables' renaming
Content-Type: text/plain; charset=us-ascii

>i.e. M200312_a changes to M0_a, M200312_b changes to M0_b ...
>     M200401_a changes to M1_a, M200401_b changes to M1_b ...
>     M200402_a changes to M2_a, M200402_b changes to M2_b ...
>     M200403_a changes to M3_a, M200403_b changes to M3_b...

A little simple code should be able to do this:

data _null_ ;
  set sashelp.vcolumn end=eof;
  where libname='WORK' & memname='ORIG' ;
  if substr(name,5,1)='3' then newname='M0' || substr(name,8,2) ;
  else newname='M' || substr(name,7,3) ;
  if _n_ = 1 then call execute ('data work.orig;set work.orig;') ;
  call execute('rename ' || trim(left(name)) || '=' ||
trim(left(newname)) || ';') ;
  if eof then call execute('run;') ;
run;

>Next month, I will need to add another month's data to each dataset( by
>adding M200404_a M200404_b M200404_c M200404_d ) and also there will be
>a new dataset base200403:

I can't help thinking, though, that you'd be much better off simply
storing the month (and year) in a separate variable and using the same
names for values in each new month, so you'd end up with variables
month (currently with values 0,1,2 & 3) , a, b, c and d.

This would be much easier to manage, as well as making any future
analyses much morer straightforward.

--
Ace in Basel - brucedotrogers a.t rochedotcom