Adjusted R-square with or without fixed effects

easygoing — Fri, 04 Nov 2016 17:15:07 GMT

Dear experts,

This question is somewhat econometrics question. I believe that experts here can help me understand this.
I'm estimating panel regressions and have two equations, which are exactly the same, except that 
in Equation 1 both time and firm fixed effects are included whereas in Equation 2 only time fixed effects are included:
Equation 1: y x1 x2 x3 timedummy firm-dummy 
Equation 2 : y x1 x2 x3 timedummy
Looking at the adjusted R-Squared I was surprised to see that it is
lower in Eq.1 than in Eq.2. As explained in econometric books, adding fixed effects is simmilar to
including dummies for each individual. Thus,
intuitively, as there are more variables in the Eq.1, I thought that
the adjusted R-squared should be higher than in Eq.2.

Why adjusted R-squared is lowed in Eq.1? It would be greatly appreciated if you help understand this!

Thank you.

Re: Adjusted R-square with or without fixed effects

ballardw — Fri, 04 Nov 2016 17:25:19 GMT

I would suspect the actual values of the Firm_dummy variable have a lot of bearing.

You should show which regression code you ran and the diagnostics output, or better the actual data and code so we may duplicate your results. You could use instructions here https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-data-AKA-generate/ta-p/258712 on how to generate data step code of your data that you can paste to the forum to recreate your data (or a subset sufficient to demonstrate the issue).

topic Re: Adjusted R-square with or without fixed effects in SAS Programming

Adjusted R-square with or without fixed effects

Re: Adjusted R-square with or without fixed effects