[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Not concave in xtfrontier

From   "Scott Merryman" <>
Subject   Re: st: Not concave in xtfrontier
Date   Thu, 12 Jul 2007 18:08:02 -0500

I don't think your specification is appropriate for a stochastic
frontier model.  If you wish to control for firm specific effects use
-frontier- with firm dummy variables rather than first differencing.


On 7/12/07, vinish kathuria <> wrote:
Thanks Scott!

If I don't suppress constant term, for few it
converges but for few it still shows - not concave.

However, the model also requires suppressing constant
term - as it is in difference!


*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index