st: Re: Creating a dataset from a file with multiple observations

["Michael Blasnik" <michael.blasnik@verizon.net>]

I'm not sure I fully understand, but this may do what you want:

bysort ID Procedure (Date): keep if Procedure==1 & Date==Date[1]

Michael Blasnik
michael.blasnik@verizon.net

----- Original Message ----- 
From: "Natarajan, Raj" <Raj.Natarajan@roswellpark.org>
To: <statalist@hsphsun2.harvard.edu>
Sent: Wednesday, December 15, 2004 10:12 AM
Subject: st: Creating a dataset from a file with multiple observations



> Dear Statalist:
>
> I have to create a data set with ID,dates and classifications for a
> particular procedure done on patients.
>
> Detail:
>
> I have a large dataset  with four variables similar to the example
> below: ID, Procedure (procedure done on different sites), Date of
> procedure (Date) and Class (Classification).
>
> ID Procedure Date Class
>
> 934510 1 27 Aug 03 1.0
> 934510 1 27 Aug 03 8.0
> 934510 1 27 Aug 03 15.0 
> 934510 2 24 Sep 03 1.0
> 941423 4 10 Oct 02 15.0
> 941423 1 10 Oct 03 13.0
> 834561 1 20 Dec 02 16.0
> 834678 1 12 Jun 01 1.0
> 834678 2 21 Feb 02 5.0
> 834678 1 14 Mar 02 18.0
> 834678 3 02 Nov 03 30.4
>
> I would like to create a data set for the first procedure 1 done on the
> same date with the classifications. The outcome should be like this:
>
>
> ID Procedure Date Class
>
> 934510 1 27 Aug 03 1.0
> 934510 1 27 Aug 03 8.0
> 934510 1 27 Aug 03 15.0
> 941423 1 10 Oct 03 13.0
> 834561 1 20 Dec 02 16.0
> 834678 1 12 Jun 01 1.0
>
> Any suggestions?
>
> Thanks.
>
> Raj Natarajan
> Roswell Park Cancer Institute
> Buffalo, NY 14263
> Phone: 716-845-8085
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/