Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Question regarding deterministic frontier analysis


From   Scott Merryman <scott.merryman@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Question regarding deterministic frontier analysis
Date   Thu, 29 Nov 2012 08:42:53 -0600

On Wed, Nov 28, 2012 at 1:00 PM,  <irene@theicct.org> wrote:
> Hi,
>
>
> I am new to STATA 12 and am trying to model a deterministic cost frontier using a set of panel data. However, I know that STATA's "frontier" function is for stochastic frontier. Is it possible to model with just a u (technical inefficiency) term rather than also a v (random noise) term (i.e. all of the variation is captured as technical inefficiency)?

Yes, by using the appropriate constraint to make sigma_v as small as
possible or use Corrected Ordinary Least Squares (COLS).

webuse frontier1,clear
*Can get somewhat close with the half-normal distribution
constraint define 1 [lnsig2v]_cons = -6
frontier lnout lncapital lnlabor, constraint(1)

*better with the truncated normal
constraint define 2 [lnsigma2]_cons = `=invlogit(1)'
frontier lnout lncapital lnlabor, constraint(2) d(tn)

*COLS
reg lnoutput lnlabor lncapit
predict residual, re
sum res
gen u = residual -r(max)
gen te = exp(u)

Scott

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index