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.

# st: Question regarding deterministic frontier analysis

 From irene@theicct.org To statalist@hsphsun2.harvard.edu Subject st: Question regarding deterministic frontier analysis Date Wed, 28 Nov 2012 11:00:00 -0800 (PST)

```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)?

So far, I've tried

frontier lnvar1 lnvar2 lnvar3, dist(hnormal) cost

Iteration 0:   log likelihood =  16.566731
Iteration 1:   log likelihood =  16.600926
Iteration 2:   log likelihood =  16.730963
Iteration 3:   log likelihood =  16.747086
Iteration 4:   log likelihood =  16.750521  (not concave)
Iteration 5:   log likelihood =  16.755428
Iteration 6:   log likelihood =  16.755633  (not concave)

Iteration 44:  log likelihood =  16.757851  (not concave)
Iteration 45:  log likelihood =  16.757851  (not concave)
Iteration 46:  log likelihood =  16.757851  (not concave)
Iteration 47:  log likelihood =  16.757851  (not concave)
Iteration 48:  log likelihood =  16.757851

Stoc. frontier normal/half-normal model           Number of obs   =         15
Wald chi2(2)    =    3672.72
Log likelihood =  16.757851                       Prob > chi2     =     0.0000

------------------------------------------------------------------------------
lnvar1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
lnvar2 |     .87537   .0853459    10.26   0.000     .7080951    1.042645
lnvar3 |   .1496319   .0825894     1.81   0.070    -.0122404    .3115042
_cons |  -5.472008   2.328865    -2.35   0.019     -10.0365   -.9075169
-------------+----------------------------------------------------------------
/lnsig2v |  -5.072258   .3681989   -13.78   0.000    -5.793915   -4.350602
/lnsig2u |  -17.64313   37500.48    -0.00   1.000    -73517.24    73481.95
-------------+----------------------------------------------------------------
sigma_v |   .0791723   .0145756                      .0551909     .113574
sigma_u |   .0001475   2.765992                             0           .
sigma2 |   .0062683   .0023471                      .0016681    .0108685
lambda |   .0018632   2.767901                     -5.423124     5.42685
------------------------------------------------------------------------------
Likelihood-ratio test of sigma_u=0: chibar2(01) = 0.00   Prob>=chibar2 = 1.000

Thank you,
Irene

*
*   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/
```