Date:         Tue, 21 Jun 2011 12:23:57 -0400
Reply-To:     Tom Abernathy <tom.abernathy@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Tom Abernathy <tom.abernathy@GMAIL.COM>
Subject:      Re: A data manupulation Question

So if that works then you can simplify the statement to just

  if . < stdat <= datstop then delete;


On Tue, 21 Jun 2011 12:15:26 -0400, Tom Abernathy <tom.abernathy@GMAIL.COM>
wrote:

>From your original example data you seem to be asking for this statement:
>
>if nmiss(datstop,stdat) = 0 then
>  if not (datstop < stdat) then delete
>;
>
>This statement can be read in English as
>
>When both DATSTOP and STDAT have values then delete cases where DATSTOP is
>not less than STDAT.
>
>If this doesn't work for your complete data then update your example with
>the cases where you want a different rule.
>
>On Tue, 21 Jun 2011 11:59:39 -0400, Tom Smith <need_sas_help@YAHOO.COM>
>wrote:
>
>>On Tue, 21 Jun 2011 11:54:20 -0400, Tom Smith <need_sas_help@YAHOO.COM>
>>wrote:
>>
>>>You are correct but also note that this problem is little more
complicated
>>>than any of you thought. plz find he two conditions below:
>>>
>>>1) if variable "STDAT" is not missing then keep the observations where
>>>   ----------------------------------
>>>DATSTOP < STDAT.
>>>
>>>Note: I think you both missed the part "if variable "STDAT" is not
>>>missing". here only talig about "STDAT", NOT bother "STDAT" and
"DATSTOP".
>>>
>>>2)if variable "STDAT" or "STDAT" is missing then the observations
remains.
>>>
>>>Note: you are right any of them can be missing.
>>>
>>>Please help me. I am cofussed. Also nmiss fuction is not working for me.
>>>looks like I have a old sas version. Please make if compatable to old
SAS.
>>>
>>>Thank you so much from the bottom of my heart.
>>
>>Nmiss fuction is working. But the other codes for the conditions are not
>>quite correct. not givingthe correct result. can anyone please be kind
>>enough to look at the problem very carefully? Thanks. I am lost.