Hello,
In comparing regression results with a colleague running a tobit model (as treatreg) in SAS, we've discovered that although the linear (xb) prediction is identical for both programs, other predictions are different, such as e(a,b), and ystar(a,b).
I am interested in calculated the predicted value from tobit that would represent both the probability of being in the uncensored part of the distribution and the effect of the explanatory variables.
In SAS, this prediction is calculated manually as using:
CDF(xb/sigma)*(xb+sigma*lambda) [note: I believe that lambda is the inverse Mills ratio]
Any direction on the equivalent prediction mechanism in Stata would be much appreciated. On a related, I'd like to be able to calculate the marginal effects relative to this prediction as well.
Thanks!
Frances Ames
***********************************************************
Warning: This email may contain confidential or privileged information
intended only for the use of the individual or entity to whom it is
addressed. If you are not the intended recipient, please understand
that any disclosure, copying, distribution or use of the contents
of this email is strictly prohibited.
***********************************************************
*
* 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/