does anybody have an idea on how to estimate a system of three seemingly
unrelated regressions, two of which being TOBIT and one OLS?
Background: I am currently estimating a translog cost function and two
corresponding cost share equations using systemfit and the seemingly
unrelated regression ("SUR") specification.
However, I consider it more appropriate to estimate the cost function
via OLS and the cost share equations via TOBIT due to some peculiarities
of the data (but still as SUR).
I am looking forward to your replies!!
(I have checked on various help sites but found nothing that could be
applied to this specific problem.)
(I also posted on stackoverflow since I do not know about the overlap of
Dr. Thomas Kopp
The University of Göttingen
Department of Agricultural Economics and Rural Development
Chair of Agricultural Market Analysis