Re: [Gretl-users] Limited Number of Independent Variables - Random Effects - Panel Data
On Wed, 13 May 2009, Nicolai C. Striewe wrote:
I have to add to my recent request that the independent
variables that I want to add to the existing 54 independent
variables are unit-invariant (only time varying) macroeconomic
variables. Only if they are unit-invariant one of the
time-dummies is omitted. It actually does not omit the time
dummy when I add a time-varying and unit-varying variable. My
question is why is a time dimension/time dummy omitted as one
unit-invariant time-varying variable is added?
I suggest the same answer as before: exact collinearity. If you
believe that is not the case, please post an example and we'll
take a look. (You could test this hypothesis by running a
regression of the "newly added" variable on all of the original
regressors, including the time dummies.)
Allin Cottrell