Dear Statalister,
I have a question on -dtobit- which I've seen it was already asked but
nobody answered. I can't understand what exactly calculate -dtobit- with the
option -censor-. Is it a particular form of the -at(matname)- option where
values are those of the uncensored sample or not? I guess it must be a
different stuff because if I ask for marginal effects at the means of IV for
the uncesored sample I do not get the same results as with the -censor-
option. Could anybody tell me what is the actual meaning of this option?
thanks
Renzo
--
Renzo Carriero
Dipartimento di Scienze Sociali
Università degli Studi di Torino
via S. Ottavio 50, 10141 Torino
tel.: +39 011 6702658
fax: +39 011 6702612
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
