turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- Base SAS Programming
- /
- Adjusted R-square with or without fixed effects

Topic Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

11-04-2016 01:15 PM

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.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to easygoing

11-04-2016 01:25 PM

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-dat... 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).