near singular matrix error (dummy variable trap)

#1
Hi,

Im running a regression in eviews and have 11 dummy variables for 12 different industries. As i already dropped one i do not understand why i am reviewing a "near singular matrix error" (eviews) indicating perfect colinearity . Futher, I eliminated another 4 dummy variables to finally get to an result. Any ideas what i can do about that if i want to control for the 12 industries?

Thank you very much in advance
 

Lazar

Phineas Packard
#2
Something is wrong here. Could you give more information on the model. Is it possible that you accidently have a linear combination of variables in your model somewhere.