Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Fixed effect logit with sampling weights (pweights)

From   "Jesper B. Sorensen" <>
Subject   Re: st: Fixed effect logit with sampling weights (pweights)
Date   Wed, 06 Apr 2005 07:29:00 -0400

Hej Michael,

I wanted to do this once as well, but it is not possible b/c xtlogit, fe does not accept weights.

However, clogit does accept weights - but only iweights (importance weights). It is not easy to determine how those weights are incorporated, but a query to Tech Support generated the following:

When you specify weights the formula for the conditional likelihood in [R]
clogit p.186 is just a weighted sum of the group likelihoods with weights
being constant over the group, i.e. weights are just group likelihood
multipliers. In clogit, iweights are basically treated just as fweights,
except that they can be non-integer. The only other difference between
fweights and iweights is that fweights affect the displayed sample size,
iweights do not.
Hope this helps.


At 09:41 AM 4/6/2005 +0200, you wrote:

I am new to the list, so apologies if this issue has been discussed 100
times before. I could not find anything on it in the archives, though.
The question is:

Does there exist a procedure in STATA which allows you to estimate a
fixed effect logit model using pweights? xtlogit with fixed effects does
not allow pweights.

Thanks in advance for any help.
Michael Rosholm :-)
Ph.D, Professor,
Department of Economics,
University of Aarhus
Phone: +45 89 42 15 59  Fax :  +45 86 13 63 34

*   For searches and help try:
Jesper B. Sørensen
Richard S. Leghorn (1939) Associate Professor of Strategic Management
Sloan School of Management
Massachusetts Institute of Technology
Cambridge, MA 02142
(617) 253 7945 -- voice
(617) 253 2660 -- fax

* For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index