Dear statlisters:
The fact that dtobit provides marginal effects for the intercept term is
something which I cannot explain. In my view marginal effects should be
calculated for all covariates except for the constant term. In fact, I
don't know how to interpret the marginal change of something (intercept)
which is fixed by its very nature. May be I am missing something...
I would also like to note that <mfx compute> omits the partial effects
for the intercept from its output, and I think it would be desirable to
have dtobit behave similarly.
I would appreciate any comments on this.
Giovanni
----------------------------------------------------------
Giovanni VECCHI
Dipartimento di Economia e Istituzioni
Universit� di Roma "Tor Vergata"
Via Columbia, 2
00133 Roma
Italy
voice + 39 (06) 7259 5730
fax + 39 (06) 2020 500
email [email protected]
webpage http://www.economia.uniroma2.it/dei/membri/vecchi
----------------------------------------------------------
*
* 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/
