unobserved heterogeneity [was: Re: st: Markov model]

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Fri, 11/3/11, Mauricio Esteban Cuak wrote:
> I'm interested in knowing the reasons for which you
> assert that one shouldn't control for unobserved
> heterogeneity.

(Warning: It is Friday afternoon here and I feel like
provoking.)

Because we are doing empirical research. The strenght
of our argument comes from the fact that we base our
conclusions on stuff we have seen and not on stuff
we have imagined. The problem with unobserved
heterogeneity is that it is unobserved, so any
"solution" will be strongly relient on "imagined facts"
rather than real observations, thus dimishing the
strength of our argument.

This is a bit tongue in cheek, but I do think that
there is a core of truth in it. Actually, I think it
is entirely true, only a bit one-sided. Unobserved
heterogeneity can be a legitimate problem to worry
about, but the question is, can there be a solution?

A more balanced discussion of my position on this issue
can be found here:
<http://www.stata.com/statalist/archive/2011-03/msg00231.html>

and here:
<http://www.stata.com/statalist/archive/2011-03/msg00255.html>

-- Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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/