[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: what T/N relation in sureg?
> I am trying to conduct a SUR on a panel dataset using
> -sureg-, with one
> equation per country. However, unless I use only a very
> small number of
> equations, I always get the error message "insufficient
> observations", even
> if T > 2M (time dimension more than twice as large as
> number of equations).
> It is quite obvious that T has to be larger than M, but by
> how much? I could
> not find details on this question anywhere. Does maybe
> anybody of you know?
> (Or is it maybe so obvious that it should jump to my eyes?)
The criteria behind this appear to emanate from -reg3-.
Examination of documentation or the source code, reg3.ado,
might help here.
* For searches and help try: