[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Simultaneous Panel Data model
I want to estimate a simultaneous model with panel data using 2sls.
The model is as follows (I have omitted time and individual
subscripts, as well as errors):
y1= a + bx1 +bx2 + ... + by2
y2= c + bz1 +bz2 + ... + by1
This would be the kind of model for which one would use -reg3- if
we were talking about cross section data.
Now, the problem is that if I estimate each equation separately,
using the two-stage procedure, estimates will be inefficient when
compared to the system estimates.
Does anybody have any ideas as to how I might solve the problem?
I would appreciate any insights you may have.
Thanks in advance,
* For searches and help try: