Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Power law or Pareto distribution


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Power law or Pareto distribution
Date   Tue, 6 Mar 2007 12:54:06 -0500

Patrick Wöhrle Guimarães--
Looking briefly around in Stata, I have not seen an estimator for the
parameters of a power law distribution, and I can see how it might be
hard to write one that you had faith in, since some power
law-distributed variables have no finite mean or variance.  Perhaps
you could instead specify the distribution as lognormal?  This paper
discusses these two related families of distributions:
http://stat-www.berkeley.edu/users/aldous/Networks/1089229510.pdf

Does anyone on the list know of a method to test whether a lognormal
or power-law distribution fits the data better?  It seems like such a
method would have broad applicability to some of the arguments in that
paper...

On 3/4/07, Patrick Wöhrle Guimarães <probabilidade@gmail.com> wrote:
  I would like to estimate a power law (or Pareto distribution) for a
large dateset with complex design (income distribution) using
pseudo-log likelihood estimators. I tried to check if there is a
do-file for estimating Power law but i didnt find. Is there anybody
that did this for a large dataset with complex design? Best regards...
*
*   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/



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