Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: Hausman test and interaction


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Hausman test and interaction
Date   Mon, 17 Dec 2012 15:07:04 -0000

Dear Statalisters:

I received the following email from Niels Botterill ( botterill87 at
hotmail.com ).  He wanted to post this to the list, tried to s*bscr*b*
but failed for technical reasons beyond his control.  If anyone has any
advice I am sure he would be grateful to hear about it on the list.

--Mark

NB: If you are wondering about the "s*bscr*b*" ... I tried several times
to post this to Statalist, and the list server complained about the
content.  If this attempt gets through, I will know what the list server
didn't like.

<snip>

I have have specified a RE model (data are 35.000 respondents in 20
countries), which did not get rejected by the Hausman test. That was
good news, but when I include a cross-level interaction term between an
individual level variable and a country level variable that were both
allready in the model with main effects, the hausman test rejects my
RE-model. I don't really get how this is possible, but I havent been
able to find much on testing the RE-specification, when having
cross-level interaction terms included. Is the test still valid, and if
not what can be done instead?

Also, I have been wondering how it's possible to have valid RE-model,
that get's invalid (due to the hausman test), when I include more
country-level variables. How can my individual level variables be
unbiased, when a variable is omitted, but become biased, when it is
included?



----- 
Sunday Times Scottish University of the Year 2011-2013
Top in the UK for student experience
Fourth university in the UK and top in Scotland (National Student Survey 2012)


We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.

Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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