Re: st: how to assign new identifier numbers with duplicates

["Linda, X Zhu" <lx_zhu@hotmail.com>]

Dear Svend,

That's really great! Thank you so much for the demenstration of building new 
id combining the region (or subregion) with firm number. I am learning Stata 
and you really helped me.

Linda



> From: "Svend Juul" <SJ@SOCI.AU.DK>
> Reply-To: statalist@hsphsun2.harvard.edu
> To: <statalist@hsphsun2.harvard.edu>
> Subject: Re: st: how to assign new identifier numbers with duplicates Date: 
> Wed, 28 Mar 2007 09:35:55 +0200
>
> Linda wrote:
>
> I am using a firm-level panel data for performance analysis
> of firms. But I found that my dataset has duplicated identifiers
> (e.g. the same identity numbers for two firms in two different
> regions in a certain year). My dataset looks like as follows
> (a1 is a code for high-level region, a2 is the subregions, a3
> is the firm identity number):
>
> year a1 a2 a3
> 1995 450 57 206141
> 1995 450 54 206141
> 1996 450 57 206141
> 1996 450 54 206141
> 1997 450 57 206141
> 1997 450 54 206141
> 1995 470 41 223243
> 1995 470 43 223243
> 1995 470 44 223243
> 1996 470 41 223243
> 1996 470 44 223243
> 1997 470 41 223243
> 1998 470 41 223243
> 2000 470 41 223243
>
>
> This moment, I don't want to consider the differences of subregions.
> So, I want to change the identity number such that I have uniquely
> identified observations by the identifier variable a3 and year....
> ----------------------------------------------------------------
>
> I am not sure what you want. You tell us that two different firms
> in two different regions (a1) can have the same id number (a3). You
> then consider it a problem that the same firm id occurs in several
> times in a given year. From you sample data it seems that this
> occurs because the same firm has a record for each year and subregion,
> but you don't want to consider subregions.
>
> If the problem is, as you describe, that the same firm id is used for
> different firms in different regions, you could combine firm id (a3)
> and region id (a1) to get a unique firm id. But first a couple of
> warnings.
>
> 1) In your attempts you -replace-d the original firm id by a modified
> id,
> thus destroying the original information. This is dangerous behaviour.
>
> 2) In long id numbers you may get precision problems; use string
> variables
> to prevent that (see http://www.ats.ucla.edu/stat/stata/faq/longid.htm).
>
> Here I construct the string variable -newid-; for the first observation
> in the sample data it becomes "206141-450":
>
>    generate sa1=string(a1,"%03.0f")
>    generate sa3=string(a3,"%6.0f")
>    generate newid=sa3 + "-" + sa1
>
> Hope this helps
>
> Svend
>
> __________________________________________
>
> Svend Juul
> Institut for Folkesundhed, Afdeling for Epidemiologi
> (Institute of Public Health, Department of Epidemiology)
> Vennelyst Boulevard 6
> DK-8000  Aarhus C, Denmark
> Phone: +45 8942 6090
> Home:  +45 8693 7796
> Email: sj@soci.au.dk
> __________________________________________
>
> *
> *   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/
_________________________________________________________________
Play online games with your friends with Messenger 
http://www.join.msn.com/messenger/overview

*
*   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/