EViews.com

I am using panel data of some 60 countries over a 30 year period. One of my regressors is the size of the economy of a country which I divide into high and low income groups. I created a series for "size" where high income countries get a value of one and low income ones get a value of zero. The file is attached. When I try to estimate Eviews returns the message "near singular matrix" and won't estimate it. Can you help please?

You've probably hit the dummy variable trap.

May I know if there any diagnostic test in panel data that need to apply after estimation so tat the result is robust?

Hi Gareth:

So what should I do to avoid the trap?