st: Manual calculation of scale factors after tobit and/or truncreg [econometrics help]

[Clara Mukuria <C.Mukuria@sheffield.ac.uk>]

Hi,

I am trying to work out how to calculate the scale factor used to adjust 
results after a tobit and/or a truncreg regression. i am using tobit to 
map a condition-specific measure (several dimensions/items as 
independent variables) onto a single generic measure which takes values 
from -0.59 to 1.

I know there are several post estimation options, I am using the 
'predict y, ystar(ll, ul)' option to assess how well the the model 
predicts. However, the aim is to provide generate a mapping function 
that can be used in other datasets to predict the single generic measure 
so I need to work out the scale factor. I have seen different examples 
for the marginal effects - which is what most studies are interested in 
- but nothing for the option I am using. So my questions:

- is the post estimation option I am using appropriate?
- if it is, how do i calculate the scale factor so that other users can 
use it to predict values?
- if it isn't, what is the most appropriate approach?

I also use truncreg in a two-part model - here the data is truncated 
(excluding both y and associated independent variables). Again, I need 
to work out the scale factor.

Any help or pointers are appreciated.

Thanks,
Clara
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/