Statalist


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

Re: st: Interaction terms in fixed effects analysis


From   tara.iyer@duke.edu
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Interaction terms in fixed effects analysis
Date   Sun, 8 Feb 2009 08:58:13 -0500

Thank you. The regression results are slightly puzzling to me as well. However,
it could be because money is not allocated efficiently, or becase of
significant teacher absenteeism (a problem in India).

I have a few questions based on your answers, and would be grateful if you would
look over them.

1. Why would one subtract the coefficient from the regression constant in
analyzing the effect of expenditure in State 1 ie. 97.34217
-0.0024748*perstudent?

2. The thing is, I am estimating a different set of dummies with xi (I had to
create the variable state with State 1 = 1 and so on), and _Istate_2 and
_Istate_3. Given this, would one still subtract the coefficient on perstudent
from the constant fro State 1? Similarly for State 2?

Thanks so much.

TI



Quoting "Kyle K. Hood" <kyle.hood@yale.edu>:

Tara-
1. In OLS, is the effect of expenditure in State 2 on *testscore* 11.16 units more than State 1? Or is it 11.16 - 0.0068 (-0.0068 is the coefficient on the
interaction between state 2 and expenditure)

For state 1, you have this model: 97.34217 - 0.0024748*perstudent
For state 2, you have this model: (97.34217+11.15724) + (-0.0024748-.0067742)*perstudent
Et cetera
2. In Fixed Effects, _Istate_2 and _Istate_3 were dropped. _IstaXpers~2 and
_IstaXpers~3 were not. How would one interpret the coefficients on IstaXpers~2
and _IstaXpers~3 in the Fixed Effects model?

FE is basically just an OLS. Any dummies that are taken care of by the fixed effects procedure will be dropped. The other coefficients are interpreted in the same way (they should be exactly the same, and they should be interpreted in the same way). If you use fixed effects, you don't need to use xi, unless you are estimating a different set of dummies with xi.
3. Does the coefficient on *perstudent* indicate that increasing expenditure
across *all three states*  is decreasing testscores by 0.0025 units? Or does
*perstudent* refer to a specific state?

Perstudent represents the slope of state 1. Each of the others will be the sum of this coefficient and the state-specific coefficient. See above.
Thank you. I really appreciate your taking the time to help.

Do you know why scores go down as spending goes up?

--
Kyle Hood
Department of Economics
Yale University
New Haven, CT
website: http://www.econ.yale.edu/~kkh25





*
*   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