EViews.com

Hi all,
I am in trouble with fixed effects. I have run regressions on my panel data with 8 cross-sections, each of them referring to a country, and different years. I have always been able to obtain the values of fixed effects in every regression; that is, I get an effect for each of the countries in my panel (see the regression report attached).
However, a colleague has raised the following concern. She asks how,if the pooled regression is based on data from eight countries, could there be eight fixed country effects? She states that this might cause linear dependence and explode the equation. Therefore, she assumes that I have done something to be able to obtain those results.
I have no idea what she is referring to as I never suspected that I had to anything to my specifications or eviews output to obtain my fixed effects. Do you have any clue? Could you fill me in with what Eviews 6 may have been doing?
I would appreciate very much your help. Thanks.
A

Fixed effects are just dummy variables for each country. Your colleague is correct in stating that usually you cannot have a full set of dummy variables (i.e. a fixed effect for each country) and have an intercept in your equation, since that would cause perfect multicollinearity.

However, what EViews does is express each country's fixed effect as the difference from the over all intercept, rather than as an individual intercept. Doing it this way means there is no multicollinearity problem.

Thanks a lot, that helped!