Hello,
I have a fixed effects model as follows:
Yit = a + bi + cXit + et
Where: Yit = the dependent variable for case i at time t
a = the intercept
bi = the fixed effect for case i
c = a vector of coefficients
Xit = a vector of factors i at time t (independent variables)
et = the error term at time t
Is this an acceptable (OK) fixed effects model? I note that most models do
not use a vector of c coefficients and treat Xit as a scalar variable.
Also, approximately what sample size (number of time periods) would I need
to estimate this model when the vector c has 5 coefficients, and i has 6
cases. I realize it depends, but is there a "rule of thumb" for the number
of time periods needed?
Thanks (This is a question from a "rookie", but any response is
appreciated)
Roger Schroeder
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/