Thank you, Nicola. I am working with the package FRONTIER (developed by Coelli (1996)) yet. But this is a very "simple" program, if you want to test restrictions, you have to calculate it on your own. If the Battese and Coelli (1995) model would be implemented in Stata such things would be much more easier therefore I asked whether it is implemented in Stata.
Alexander
"Battese G.E. and T.J. Coelli (1995), A Model for Technical Efficiency Effects in a Stochastic Frontier Production Function with Panel Data, Empirical Economics, 20, 325-332."
"Coelli T.J. (1996), A Guide to FRONTIER version 4.1: A Computer Program for Stochastic Frontier Production and Cost Function Estimation, CEPA Working papers, No. 7/96."
>>> <[email protected]> 21.04.2007 12:23 >>>
I strongly suggest you to move to Frontier, a free package by Coelli (not sure if it is still around there: http://www.une.edu.au/econometrics/cepa.htm).
As far as I know, Battese (note the spelling!) and Coelli (1995) is “A Model for Technical Inefficiency Effects in a Stochastic Frontier Production Function for Panel Data”, Empirical Economics, 20, 325-332; thus it is a model for panel data.
The panel command for frontier analysis is -xtfrontier-, where none of the options you described exist (they are only in the cross-sectional command), and where surprisingly there are no options for the Battese/Coelli (1995)
approach.
However, option cm resembles a modified Battese/Coelli (1995) approach for cross-sectional data more than option uhet. In fact, in the Battese/Coelli (1995) approach, it is the mean of the technical inefficiency component that is modeled as a linear function of some covariates, not the variance. Moreover, the original Battese/Coelli (1995) approach uses a truncated normal distribution for the technical inefficiency, and Stata prevents you from selecting option uhet with a truncated normal distribution.
Nicola
At 02.33 21/04/2007 -0400, "Alexander Kalb" wrote:
>Hello,
>
>I want to estimate a stochastic frontier with the Batesse/Coelli
>(1995)
>approach. Now I wanted to ask if this is implemented in Stata. In the
>Statalist I read that one can use the "uhet"-option or the
>"cm"-option.
>Does one of these two options represent the Batesse/Coelli (1995)
>approach and if this should be the case which is the right one? Thank
>you!
*
* 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/
*
* 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/
