Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Question about generating the weight of WLS


From   Nick Cox <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: Question about generating the weight of WLS
Date   Thu, 25 Apr 2013 11:56:13 +0100

This is a common kind of question in data management, so you should
expect that -search- will turn up resources. Even if the keyword to
use in -search- is not evident, browsing the data management FAQs
reveals

FAQ     . . . . . . . . . . . . . . . . . . .  Number of distinct observations
        . . . . . . . . . . . . . . . . . . . . . . . N. J. Cox and G. Longton
        4/02    How do I compute the number of distinct observations?
http://www.stata.com/support/faqs/data-management/number-of-distinct-observations/

A general survey of counting distinct values is given in

SJ-8-4  dm0042  . . . . . . . . . . . .  Speaking Stata: Distinct observations
        (help distinct if installed)  . . . . . .  N. J. Cox and G. M. Longton
        Q4/08   SJ 8(4):557--568
        shows how to answer questions about distinct observations
        from first principles; provides a convenience command

A .pdf is accessible at
http://www.stata-journal.com/sjpdf.html?articlenum=dm0042

Those surveys show several ways to approach this. Here is one:

egen tag = tag(country mfi year)
egen ndistinct = total(tag), by(country year)

In fact your example data suggest that

bysort country year : gen ndistinct2 = _N

would work too, but the first method is more general, just in case
there is a good reason for duplicates.

After that you just feed that variable within an expression for
weights to -regress-.

Nick
[email protected]


On 25 April 2013 11:37,  <[email protected]> wrote:

> - I have stata IC 10
>
> -I have a panel data. To take in to consideration the considerable variation
> of the MFIs  (microfinance institutions) across countries on the empirical
> results, I resort to a weighted least squares with the weight being the
> inverse of the number of MFIs for the country in the database in a given
> year.
>
> Can someone help me on how to generate the inverse of the number of MFIs for
> the country in the database in a given year and how to handle this WLS.
> Please find below an example  of the data
> Thanks for your time and consideration
>
> year        mfi        country
> 2004        AFS        Afghanistan
> 2004        BRAC - AFG        Afghanistan
> 2004        FINCA - AFG        Afghanistan
> 2004        FMFB - AFG        Afghanistan
> 2005        AFS                 Afghanistan
> 2005        BRAC - AFG        Afghanistan
> 2005        CFA                 Afghanistan
> 2005        FINCA - AFG        Afghanistan
> 2005        FMFB - AFG        Afghanistan
> 2005        MADRAC                 Afghanistan
> 2004        ASC Union         Albania
> 2004        BESA                  Albania
> 2004        FAF-DC                  Albania
> 2004        NOA - ALB         Albania
> 2005        ASC Union         Albania
> 2005        BESA                  Albania
> 2005        FAF-DC                 Albania
> 2005        NOA -   ALB        Albania
> 2004        BMM C rdoba        Argentina
> 2004        Emprenda                 Argentina
> 2004        Entre Todos        Argentina
> 2004        FIE Gran Poder        Argentina
> 2004        Grameen Chaco        Argentina
> 2005        Alternativa 3        Argentina
> 2005        BMM C? rdoba        Argentina
> 2005        Emprenda                 Argentina
> 2005        Entre Todos        Argentina
> 2005        FIE Gran Poder        Argentina
> 2005        Grameen Chaco        Argentina
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index