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

Re: st: double random effects logit

From   "Stas Kolenikov" <[email protected]>
To   [email protected]
Subject   Re: st: double random effects logit
Date   Tue, 11 Sep 2007 11:22:01 -0500

I am quite not sure I understand what "random fixed effects" might
mean, but you might want to look into the ways of analyzing dyadic
data. There's a book about this
but I've never opened it, so I cannot give any comments on it -- if
anything, I would treat their methods rather cautiously, as direct
implantation of multilevel models advocated, as it seems from the
contents, by the authors, may not be totally appropriate. What was
convincing for me was the work by Peter Hoff of U of Washington where
he was showing how complicated the likelihoods of dyadic data really
are when you account for the "sender" and "receiver" effects, as well
as some related symmetries of the problem. See his papers and software
(in R) at, and the
specific paper is this one: doi:10.1198/016214504000001015

On 9/11/07, A D'Ignazio <[email protected]> wrote:
> Dear Statalist,
> I was wondering whether a stata code to estimate a double random fixed
> effects (in my dataset each observation relates to a pair of firms) has been
> developed.
> Many thanks,
> Alessio
> _________________________________________________________________
> Scarica GRATIS Internet Explorer 7 personalizzato MSN
> *
> *   For searches and help try:
> *
> *
> *

Stas Kolenikov, also found at

Small print: Please do not reply to my Gmail address as I don't check
it regularly.
*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index