Hello,
I am trying to perform a stochastic frontier analysis on a cross-sectional data of sample of banks in one country to try to estimate banks' cost-efficiency using frontier stata function.
I am using a translog function of total costs as a dependent value with 2 outputs and 2 input prices as regressors.
My problem is that the results of the exercise (after I predict inefficiency estimates) are questionable with the estimates not materially differing from one another (hundreths) all of them rougly equal to 10. (10,345 10,352 10,356, etc.) which is quite strange.
Are there any suggesitons on what's the reason behind this?
Any help much appreciated.
Best regards,
Alexander
*
* 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/
