Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Manzur Quader <manzurquader@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: heteroskedastic panel frontier |
Date | Wed, 9 Jun 2010 22:21:17 +0100 |
Hye, I am using stochastic frontier model to calculate agency cost of firms using Tobin's Q as dependent variable.I have got data of 1200 firms over 3 to 29 years.But,whenever I am using xtfrontier I am getting very low efficiency level(below 40%,which is very low compared to the literature) and this is probably due to heteroskedasticity.If I use frontier on the pooled data or on different cross sections by year conditional on uhet,I am getting much better result. Can you please tell me is there any way to incorporate heteroskedasticity if I want to use xtfrontier for my panel?I know that the Battesse and Coelli(1995) model can be estimated in stata,but is there any way I cam parameterize the mean efficiency in terms of z simultaneously?Or is there anything like uhet for panel frontier? Please advise. Regards, Syed. * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/