Dear Statlist,
I am using a tobit model to estimate an Engel curve. And after that I am
using the dtobit command to get the marginal effects (i am using stata
10). I am a bit of confused about the results that I find.
One of my regressors is log of income, do I need to add a special
specification to dtobit to get the correct marginal effect?
When I run a probit of the same model (I just change my LHS variable to a
binary choice - expenditure or not) and I then do an mfx, I don't get the
same result as the "marginal effect - probability uncensored" and I was
wondering why that was?
I will be very appreciative of your help because this is for my final year
project and if I can't interpret the result, I will look a bit like an
idiot...
Thank you in advance
Regards,
Camille
*
* 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/