EViews.com

Dear All,

I am estimating a panel regression with TSLS and whenever I include either fixed effects or time dummies I get the following error message: "positive or non-negative argument to function expected". Could anyone help me? Please see equation1 in the attached file (forum.wf1).

Thanks!

Marcelo

There is no equation1.

Sorry, I have now attached the correct file with equation1. The error message appears when I try to include fixed effects (either cross-section or time).

Thanks for your help.

What's going on is that your equation is generating an error in computing the J-statistic due to rank deficiency. Let me see if I can improve the handling of the error.