I don't have a direct answer for your question. But since the command has been released as STB-56: sg144; SJ4-3: sg144_1, you may find such an answer by consulting the related papers, which you can now freely (!) download from http://econpapers.repec.org/article/tsjstataj/ and http://econpapers.repec.org/article/tsjstbull/.
Nicola
At 02.33 22/05/2008 -0400, Abra Marie Jeffers wrote:
>Hello,
>
>I am trying to interpret the output of -dtobit- for a continuous, level
>variable.
>
>For example:
>If -dtobit- returns that the marginal effect at the mean of the
>probability of being uncensored is 0.002 for x1, can I say that "if x1
>increases by 10 from the mean, then the probability of being uncensored
>increases by 0.02 other things equal"?
>Similarly, if -dtobit- returns that the marginal effects at the mean of
>the conditional expectation is -1.45 for x2, do I say that "if x2
>increases by 100 from the mean, then the conditional expectation
>decreases by $145"?
>
>This seems too simple given the nonlinearities involved in the calculations.
>
>I would really appreciate any suggestions on giving the decomposed
>marginal effects an economic interpretation.
>
>Thanks,
>Abra
>*
*
* 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/
