st: prediction after y-logged oheckman regression

["Vanessa Carrion" <vcarrion@vt.edu>]

Hi,

I am running an oheckman model in which y variable is logged. I read that in
order to get an anti-logged prediction of yhat, one should proceed like
this:

1. Run the regression and predict yhat,
2. Then exponentiate the predicted yhat, and finally
3. Multiply the exponentiated predicted yhat by the mean value of residuals
from the regression.

After doing so, my yhat becomes a column of missing values. Am I doing
something wrong?

On the other hand, I read that predlog is useful for OLS regression.
However, I don't know how to implement it for an oheckman.

Thank you!

Vanessa

*
*   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/