Fixed Effects Estimation

[EconometricAarhus]

Hello.

Using EViews 7.

I have uploaded the Excel file "EViewsPanelDataFEquestion", which is a more simple version of my extensive data.

I am trying to make a panel data regression and get the problem with "Near Singular Matrix". My problem is when I do my regression including two constants (hh and ii) and including cross section fixed effects it tells me "Near Singular Matrix". The length is the same for all variables and so on. I have read some articles where they include these variables without problems. They use Stata and I use EViews. I can make the regression without effects, with period fixed effects, with period random effects and cross section random effects.

What seems to be the problem? And how do i fix it?

(I hope I'm posting it the correct place).

Best regards

[EViews Gareth]

A fixed effect is just a variable that is constant across time (i.e. it takes the same value for all time periods in each cross-section). You have variables that already do that. Thus you have perfect collinearity. This is a mathematical issue, and not an EViews one.