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 on April 23, and its replacement, statalist.org is already up and running.


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

st: Binary explanatory variable that vaires by country and country fixed effects


From   "Benedikt Link" <benediktlink@gmx.de>
To   statalist@hsphsun2.harvard.edu
Subject   st: Binary explanatory variable that vaires by country and country fixed effects
Date   Mon, 16 May 2011 16:38:32 +0200

Hello,

I have a question concerning binary explanatories and country fixed effects.

I am currently running a cross-sectional regression across 15 European countries. My key explanatory variable is a dummy variable (XY) that varies by country (The variable is 1 if certain reporting standards are mandatory in the respective country and 0 otherwise). In order to determine what characteristics of XY drive the effect, I split the variable into two groups, XY_a and XY_b (XY_a is 1 if the country fulfills a certain characteristic (e.g. above median earnings management) and 0 otherwise; XY_b is 1 if the country fulfills XY but XY_a is 0, otherwise it is 0. I also include a range of controls and country fixed effects. I run the following regression:

reg depvar XY_a XY_b controls country_*, vce(cluster countryid)

Now, my problem is, that no matter according to which characteristc I divide XY into XY_a and XY_b, always one of the coefficients/t-stats of either XY_a or XY_b is exaxcly the same across the different regressions using various defintions of XY_a and XY_b. This seems to me as if Stata has a certain degree of scaling/standardization that always brings back the coefficient of one of the explanatories to the same level and only changes all others.

Is this normal or is this a sign for something being misspecified (e.g., mulicollinearity)?

Thanks for your help! I appreciate it a lot!

Benedikt
-- 
NEU: FreePhone - kostenlos mobil telefonieren und surfen!			
Jetzt informieren: http://www.gmx.net/de/go/freephone
*
*   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   |   Site index