st: SFA analysis

[Georgeta Vidican <gvidican@MIT.EDU>]

Hi there,

I have a fairly basic question as I am learning Stata and the SFA  
method.
I want to use a Cobb-Douglas specification model to estimate the  
production frontier for household farms. I am interested in modeling  
inefficiency as a function of also other covariates besides the  
conventional production factors (land, labor, capital).

However, I am not sure I am following the steps correctly. First I  
estimated a half-normal frontier function (y function of land, labor,  
capital). However, do I have use a truncated normal distribution if I  
want to include other covariates (i.e. age, education) (i.e. frontier  
ln_land ln_labor ln_capital, dist(normal) cm(list_of_covariates))?

Also, is it normal for the results of rubust regression to be very  
different from this last run?

Many thanks.
Georgeta

*
*   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/