Page 1 of 1

negative R² with System 2SLS/3SLS

Posted: Sat Dec 13, 2008 5:05 pm
by s3n470r
Hi all,

when I estimate a system of 4 equations with 2SLS/3SLS and GMM. I get a negative R² for 2 equations and positive R² for the other 2.
WHY?

If I estimate the model with SUR, they are all positive.

Any suggestions whats wrong with my model, or recommendations how to overcome this flaw.

Best,
senny

Re: negative R² with System 2SLS/3SLS

Posted: Sat Dec 13, 2008 5:21 pm
by EViews Gareth
Negative R2 in instrumental variable estimation is not a problem, and you shouldn't worry about it.