st: replacing negative predicted values from a tobit with zeros

["thouraya triki" <thouraya.triki@hec.ca>]

Hi statalist users,
I have panel data concerning the level of risk management activities in
the firm; In order to get the RM level adopted for firm value max, I
regress the RM level on some variables proxying managerail risk aversion
and took the residuals.I estimate this first model with xttobit command.
next, I replace the negative values of residuals with 0 and reestimate a
new model with these new residuals as dependent variable, I reuse the
xttobit command.
my question is: is it correct to replace negative values with 0 and
suppose that the new model is a tobit? in case this is not correct; what
is the appropriate way to do it?

Thanks for teh help
*
*   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/