Statalist


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

Re: st: areg vs xi reg vs xtreg vs what else?


From   Dana Chandler <dchandler@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: areg vs xi reg vs xtreg vs what else?
Date   Thu, 17 Sep 2009 09:22:18 -0500

Maarten -

Thanks for your reply. Are you sure that I cannot use log_population?
I may have been unclear in my description.

I should have emphasized that my log_population is constant for each
state and refers only to the log_population before my study. Hence
it's really log_population_1940. Also, I'm then interacting it with
grade-year combinations so that for example, I estimate a coefficient
that describes the effect of log_population in 1940 on say, 6th
graders in 1952. I think I am still identified if I have that
interaction term.

Best,
Dana

On Thu, Sep 17, 2009 at 8:56 AM, Maarten buis <maartenbuis@yahoo.co.uk> wrote:
> --- On Thu, 17/9/09, Dana Chandler wrote:
>> I'm running a large fixed effects model where each
>> observation is a test score for every grade level, year,
>> and state. My outcome variable is a test score and the
>> focal independent variable is the % of HH in state that
>> have cable television. The only other variable I have is
>> log_population for the first year of my study at the state
>> level which I also want to interact with each combination
>> of grade-year.
>
> Are your fixed effects States? In that case you can't use
> your log_population variable, as that is constant within
> states. The whole point of fixed effects, as that you
> control for all variation (observed and unbserved) that
> remains constant within a state. The price you pay is
> that you cannot study what these constant variables do.
> You can do so, if you are willing to do a random effects
> model. But now you have to assume that the remaining
> unobserved state level variables are uncorrelated with
> any of the observed variables. There is a huge difference
> between disciplines in the willingness to accept this
> assumption.
>
> hope this helps,
> 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/
>

*
*   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