Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

RE: st: Ginis for negative net worth


From   philippe van kerm <philippe.vankerm@ceps.lu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Ginis for negative net worth
Date   Mon, 5 Jul 2010 15:13:17 +0200

Just to confirm this: -inequal7- simply *warns* you of the presence of zero and negative values. Gini coefficients can be computed with negative values --nothing special needs be done mechanically--, but interpretation is more hazardous, and, as you discovered, it is not bounded at 1. (To see why the Gini coefficient can go above one, realize that with negative net worth, your Lorenz curve will initially go below zero (as you cumulate negative net worth first).)

I would too consider absolute Gini measures. (The -sgini- package might be useful for this -- see http://medim.ceps.lu/stata/sgini.pdf .)

Philippe

> -----Message d'origine-----
> De : owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] De la part de Nick Cox
> Envoyé : Monday, July 05, 2010 2:43 PM
> À : statalist@hsphsun2.harvard.edu
> Objet : RE: st: Ginis for negative net worth
>
> Two points:
>
> 1. -inequal7- specifically warns you of zero or negative values. So,
> there is a strong flavour of caveat emptor to the indulgence.
>
> 2. There is no need for surmise or speculation about precisely what -
> inequal7- does as you can look at the code to find out. -viewsource
> inequal7.ado- gets you there.
>
> Nick
> n.j.cox@durham.ac.uk
>
> Nora Müller
>
> thank you very much for your advice!
>
> But inequal7 (as well as ineqdec0) says, that it uses values equal and
> smaller than zero in calculating the gini coefficient and therefor I
> thought that it already uses a special algorithm for such values so
> that
> the range of the gini coefficients still is [0;1].
>
> The article you mentioned I already read, but thank you for the advice.
>
> So finally I'll try to work with the absolte gini now.
>
> Am 05.07.2010 13:01, schrieb Abdel Rahmen El Lahga:
>
> > Gini index as well as other inequality indices are defined for
> > positive outcomes and can not applied for negatives observations.
> > You have two alternatives:
> > 1) use a trick (not so popular) proposed by Chen et al. (1982) in
> > oxford economic paper (1982) page 473-478 (See also the comment by
> > Berrebi and Silber in the same journal (1985 page 525-526) to deal
> > with negative outcomes
> > 2) You can use absolute gini index to avoid problems caused by
> > negative income (but you should care about  different ethical
> > judjements associated to absolute view of inequality)
>
> > 2010/7/5 Nora Müller<nora.mueller@uni-bamberg.de>:
>
> >> I'm trying to calculate the gini coefficient for net worth (=
> financial
> >> assets + real assets - debts) that has some negative values.
> >> I do this using the command "inequal7" written by Philippe Van Kerm.
> >>
> >> I'm doing an international comparison and for Poland I get a Gini
> that is
> >> 1.15. Can anyone help me with the interpretation?
> >> Why can there be a gini greater than one?
>
> *
> *   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index