Jaren Pope sent this to me privately, and I take
the liberty of forwarding it to Statalist.
There is published work on Singh-Maddala and Dagum
distributions. I'm aware of unpublished work (not
mine) on lognormal distributions.
(I would have thought that the mean, sd and
quartiles of the fitted distribution would normally
be close to that of the observed distribution,
irrespective of the method of fitting, moments,
maximum likelihood, or something else. If they
are not, the distribution is probably a lousy fit
anyway.)
[email protected]
> I am trying
> to fit a lognormal distribution (or perhaps another reasonable
> distribution) to an income distribution I got from the US
> census. So,
> basicly I have counts of households in each income
> category. Do you know
> of a program in stata that would do this so I can
> ultimately get the mean,
> standard deviation and quartiles from the fitted distribution?
>
*
* 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/
