I think that programs svylorenz, inequal, calculate inequality measures
directly from the data, for example, a sequence of observations of income. I
want to use the DFL method to decompose the effects of the minimum wage and
the characteristics of workers. For this, obtained through this method is a
counterfactual Kernel density function. After obtaining this function, I
desire from the same function (and not directly from my observed income
data) to obtain some measures of inequality, as the difference between the
ninth decile and the first decile, the Theil index and the Gini index, as is
done in the Dinardo, J., NM Fortin, and T. Lemieux (1996) article. For the
first measure (difference between deciles) and I think also for the other
two, is necessary to obtain the cumulative distribution function related to
the Kernel density function. For this I need a procedure in Stata to
integrate the kernel density function. I discovered that in the set of
commands named moremata there is a code (mm_kint()) that apparently
integrates Kernel density functions. This is the right path?
Regards
Henrique Neder
-----Mensagem original-----
De: [email protected]
[mailto:[email protected]] Em nome de Austin Nichols
Enviada em: quinta-feira, 9 de outubro de 2008 18:05
Para: [email protected]
Assunto: Re: st: estimating distribution functions from Kernel densities
Henrique Neder <[email protected]> :
I think you need to be more clear about what you mean to do. The DFL
approach reweights the data to obtain the counterfactual density. If
you are interested in an inequality measure for the reweighted
distribution, can't you just compute the measure using your new
weights, using e.g. a program by Stephen Jenkins et al. (findit
svygei, findit ineqdeco)?
DiNardo, J., N.M. Fortin, and T.Lemieux (1996) "Labour Market
Insitutions and the Distribution of Wages, 1973-1992: A Semiparametric
Approach," Econometrica, 64(5): 1001-1044.
On Thu, Oct 9, 2008 at 4:52 PM, Henrique Neder <[email protected]> wrote:
> Dear Stata listers
>
>
> I am studying the methodology of decomposition of Dinardo, Fortin and
> Lemieux, known as Dfl. I need to estimate the inter-quantile deviation,
the
> Gini index and the Theil index for Kernel density functions. There would
be
> some method (for example, numerical integration) to do this using Stata?
>
> Henrique Neder
*
* 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/
