Date:         Tue, 20 Nov 2001 14:05:00 +0000
Reply-To:     John Whittington <John.W@MEDISCIENCE.CO.UK>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         John Whittington <John.W@MEDISCIENCE.CO.UK>
Subject:      Re: Missing Values
Comments: To: "Tift, Brian" <bet5@CDC.GOV>
In-Reply-To:  <E166B7n-0001RA-00@relay1.netnames.net>
Content-Type: text/plain; charset="us-ascii"; format=flowed

At 08:38 20/11/01 -0500, Tift, Brian wrote:

>Thanks to everyone for the suggestions, the main problem was that
>there was a long series of variables (~50) that the user wanted to run.
>The solution we came up with was from off the list,
>not the cleanest solution but it works.
>
>data temp; set test;
>array test _numeric_;
>  do i=1 to dim(test);
>   if test(i) in (888,999) then test(i) = .;
>  end;
>proc print; run;
>
>proc means data=temp; run;

Brian, that's certainly the 'most obvious way' - but unless you really
want/need to (or are using this 'modified dataset' repeatedly), there is no
actual need to create your dataset 'temp' - with just a very slight
addition to your code, you could make it a 'view' of the original dataset:

data temp / view = temp ;
    set test;
    array test _numeric_;
    do i=1 to dim(test);
       if test(i) in (888,999) then test(i) = .;
    end;
run ;
proc print data = temp ; run;
proc means data=temp; run;

Kind Regards


John

----------------------------------------------------------------
Dr John Whittington,       Voice:    +44 (0) 1296 730225
Mediscience Services       Fax:      +44 (0) 1296 738893
Twyford Manor, Twyford,    E-mail:   John.W@mediscience.co.uk
Buckingham  MK18 4EL, UK             mediscience@compuserve.com
----------------------------------------------------------------