Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: nonconvergence of switching probit when data weighted


From   Steven Samuels <ssamuels@albany.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: nonconvergence of switching probit when data weighted
Date   Thu, 12 Mar 2009 10:04:45 -0400

There is no problem, Paula. I should have looked at your output more closely. The weights were standardized so that they sum to the sample size, not the population size. This practice was common before survey software became widely available.

-Steve

On Mar 12, 2009, at 9:52 AM, Paula Albuquerque wrote:

The weights are obtained through the European Social Survey database, I suppose many researchers before me have used them. But, I have also wondered that if in Stata, the weights are the inverse of the probability that the observation is included, and probabilities should be between zero and 1, the
inverse should have a minimum of 1, shouldn't it?

Paula


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Steven Samuels
Sent: quinta-feira, 12 de Março de 2009 13:14
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: nonconvergence of switching probit when data weighted


Most of your weights are < 1, which is a sign of trouble, though
maybe not yours.  These observations do not even represent themselves.

-Steve

*
*   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/



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