> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Daniel Mueller
> Sent: 20 January 2004 15:10
> To: [email protected]
> Subject: st: RE: survey data and inequality
>
>
> > It is quite common to use survey data to compute income
> inequality and
> > poverty measures. Usually survey data is stratified so that the
> > probability of being surveyed varies across households. The
> -pweight-
> > and -svy- commands take the "expansion factors" into
> account, however
> > non of the income-distribution/poverty commands (ineqdeco, inequal,
> > rspread, sumdist or
> > povdeco) allow for the -pweight- option and I don't know how
> > to use them within the -svy- command.
> >
> > How can I use the inequality/poverty commands when I have a
> stratified
> > survey (the survey already contains the inverse of the
> probability of
> > being surveyed [expansion factors])?
>
> -findit poverty- and -findit inequality- will point you to
> several user-written programs, which might be helpful.
The programs cited will provide the correct /point/ estimates with svy
data -- just refer to your pweight as an aweight in e.g. -ineqdeco- or
-inequal7-. The programs don't produce correct standard errors, however
(or, indeed, any standard errors). -geivars-, based on Cowell (J
Econometrics 1989) producing estimates of both inequality and sampling
variances, allowing for household level clustering (but not PSU
clustering) and weights.
I am developing programs that estimate sampling variances for
generalised entropy and Atkinson indices, that optionally allow for
clustered and stratified data. (Please don't ask me for them; I'll
announce them on this list when they are ready for distribution.)
Stephen
-------------------------------------------------------------
Professor Stephen P. Jenkins <[email protected]>
Institute for Social and Economic Research
University of Essex, Colchester CO4 3SQ, U.K.
Tel: +44 1206 873374. Fax: +44 1206 873151.
http://www.iser.essex.ac.uk
*
* 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/
