Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Marginal Effects in Tobit witih Interactions


From   Hoi Kit Lo <hoikitlo@cityu.edu.hk>
To   StataList <statalist@hsphsun2.harvard.edu>
Subject   st: Marginal Effects in Tobit witih Interactions
Date   Tue, 09 Jun 2009 17:16:01 +0800

Dear all,

I am working with a Tobit with interactions in the following form:

tobit y x1 x2 x1x2, ll(0)

I want to evaluate the marginal effects on Y conditional on exceeding 
the lower end and the unconditional change in the observed Y, which
if i am correct, the STATA command for its computation is 
[mfx, predict(e(0,.))] and [mfx, predict(ystar(0,.))], respectively.

In the achieve, I was only able to find procedures to manually calculate the 
marginal effects on the probability of being uncensored [mfx, predict(pr(0,.))].

I have 2 questions:

1) Are the marginal effects reported in STATA using the above commands 
    valid when there are interaction terms in the regression?
2) If not, are there any suggestions for procedures in calculating them manually?

I have referred to a few econometrics books and the Internet but
failed to find any thing on this topic. It will be of great help if anyone
could answer this inquiry. Thank you for reading.

Hoi Kit Lo

City University of Hong Kong
Department of Economics and Finance
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index