Re: st: RE: merge

["Fabian Schönenberger" <sch.f@gmx.ch>]

Dear Simon

Sorry, I did not see your reply! Many thanks for your feedback. However, I still could not solve the problem. But, I have checked my key-variables in the master- and using-file. Cusip consists of numbers and letters; each company has its specific cusip. I encoded this variables in order to use it in regressions etc. But I think this is now a problem for mergeing data. However, if I decode the cusip again, I can not use it with xtset because its a string variable. 

Do I have to generate a new numeric variable for each cusip (without letters) in order to merge?

Fabian

-------- Original-Nachricht --------
> Datum: Fri, 1 Jun 2012 15:05:13 +0000
> Von: Simon Falck <simon.falck@abe.kth.se>
> An: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
> Betreff: st: RE: merge

> Dear Fabian, 
> 
> It is difficult to give a specific reply when you do not tell us more
> about your datasets and key variables. However, here are some general inputs on
> merging files in Stata that perhaps are useful for you.
> 
> The -merge- command enables merging files with common id´s. One-to-one 
> -merge 1:1- implies that the identifiers (key variables) are exactly the
> same in both files. If this is not your case then you should consider
> Many-to-one -merge m:1- or One-to-many -merge 1:m-. It depends on how your datasets
> are structured and their content. The two latter options are used when you
> have a common id in both files and one file, either the master or user,
> differ in for example time period. If you have two datasets which includes
> id´s that are non-common in both directions you use the Many-to-many -merge
> m:m- option.
> 
> As understood the key-variables are important to inspect according to the
> options described above. Don’t forget to inspect the format, for example
> are both key variables in string format? Consider how a key-variables
> should be constructed and what a common attribute implies and the directions.
> 
> You get a great instruction by typing: -help merge-
> 
> Good luck,
> Simon
> 
> 
> 
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of "Fabian Schönenberger"
> Sent: den 1 juni 2012 14:28
> To: statalist@hsphsun2.harvard.edu
> Subject: st: merge
> 
> Dear Statalist
> I try to merge two datasets. In both files each observation is uniquely
> identified by cusip and year. I sort both files with xtset cusip year.
> Afterwards, I command: 
> 
> merge 1:1 cusip year using "C:\Users\User\Documents\Uni
> SG\Doktorat\Data\Price Data\pricedatev5.dta", keepusing(capm_marketpremium), keep(3)
> 
> I am only interested in those observations of cusip-year-combinations
> which are in my masterfile - therefore keep(3). 
> 
> However, either the matched observations nor the unmatched observations
> are correct, meaning that I get for each cusip-year observations the wrong
> capm_marketpremium and I do not get observations in my masterfile for
> cusip-year observations although the using file has an observation. 
> 
> I also tried merge m:m but it did not work. What am I doing the wrong way?
> 
> Many thanks for suggestions.
> 
> Fabian 
> -- 
> NEU: FreePhone 3-fach-Flat mit kostenlosem Smartphone!                    
>              
> Jetzt informieren: http://mobile.1und1.de/?ac=OM.PW.PW003K20328T7073a
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

-- 
NEU: FreePhone 3-fach-Flat mit kostenlosem Smartphone!                                  
Jetzt informieren: http://mobile.1und1.de/?ac=OM.PW.PW003K20328T7073a
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/