----- Original Message -----
From: "Stephen P. Jenkins" <[email protected]>
To: <[email protected]>
Cc: <[email protected]>
Sent: Thursday, January 30, 2003 6:04 AM
Subject: st: Re: Re: gini from grouped data
> On Thu, 30 Jan 2003 05:49:14 -0600 Scott Merryman <[email protected]>
> wrote:
>
> > > ------------------------------
> > >
> > > Date: Tue, 28 Jan 2003 11:37:02 -0600
> > > From: "Mito" <[email protected]>
> > > Subject: st: gini from grouped data
> > >
> > > Dear friends,
> > > is there an ado file for the estimation gini from grouped data?
> > > thanks,
> > > Maximo
> >
> >
> > Dr. Martin Biewen has an ado file for the "Bootstrap for inequality
> > decomposition by subgroup" avaiable at
> >
http://www.uni-heidelberg.de/institute/fak18/institutes/awi/ls_schmidt/cv_bie/st
> > ata.html
> >
> > It computes additive decompositions of MLD and Theil by subgroup and
appropriate
> > bootstrap confidence intervals.
>
>
> IMHO Scott's suggestion is not what was originally asked for, namely a
> program for calculating inequality indices from grouped data (i.e. you
> know category boundaries and the numbers of obs with incomes falling
> within the categories, and maybe some category means).
> Martin's program, like my -ineqdeco-, uses micro-data (unit record
> data), allowing decompositions by /population/ subgroup.
> As Philippe Van Kerm suggested, you can use any of the microdata
> inequality estimation programs around (his, mine, ...) to calculate
> inequality measures from grouped data (using frequency weight options),
> at the cost of under-estimating inequality ... you will be assuming
> equality within income categories.
> For a review of ways to do better, I recommend /Measuring Inequality/,
> second edition, by Frank A Cowell, and the references therein
> (including Cowell and Mehta, Review of Economic Studies, 1983?)
>
> Stephen
Stephen,
Thanks for clarrifying the issue.
The complete third edition of /Measuring Inequality/ is aviable at:
http://darp.lse.ac.uk/frankweb/Frank/mi3.html#mi3.html#
Scott
*
* 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/
