Error Components Frontier (see B&C 1992)
The model is a production function
The dependent variable is logged


the ols estimates are :

                 coefficient     standard-error    t-ratio

  beta 0         0.10481463E+02  0.34880367E+00  0.30049750E+02
  beta 1        -0.46367553E+00  0.48584883E-01 -0.95436172E+01
  beta 2         0.97512202E+00  0.54915135E-01  0.17756890E+02
  sigma-squared  0.17531871E+01

log likelihood function =  -0.63399166E+04

the estimates after the grid search were :

  beta 0         0.12118757E+02
  beta 1        -0.46367553E+00
  beta 2         0.97512202E+00
  sigma-squared  0.44325098E+01
  gamma          0.95000000E+00
   mu is restricted to be zero
  eta            0.00000000E+00


iteration =     0  func evals =     20  llf = -0.27959529E+04
     0.12118757E+02-0.46367553E+00 0.97512202E+00 0.44325098E+01 0.95000000E+00
     0.00000000E+00
gradient step
iteration =     5  func evals =2051489  llf = -0.15271766E+04
     0.39643006E+01 0.52404830E+00 0.34684785E+00 0.20318046E+01 0.97573765E+00
     0.68442704E-02
iteration =    10  func evals =2051543  llf = -0.12304332E+04
     0.36593323E+00 0.91173315E+00 0.22692854E+00 0.27951545E+01 0.97838366E+00
    -0.67680118E-02
iteration =    15  func evals =2051628  llf = -0.12303208E+04
     0.35796557E+00 0.91372695E+00 0.22307827E+00 0.28537199E+01 0.97884135E+00
    -0.65288667E-02


the final mle estimates are :

                 coefficient     standard-error    t-ratio

  beta 0         0.35796557E+00  0.18079788E+00  0.19799212E+01
  beta 1         0.91372695E+00  0.22593781E-01  0.40441524E+02
  beta 2         0.22307827E+00  0.21029979E-01  0.10607632E+02
  sigma-squared  0.28537199E+01  0.17808673E+00  0.16024326E+02
  gamma          0.97884135E+00  0.14380644E-02  0.68066588E+03
   mu is restricted to be zero
  eta           -0.65288667E-02  0.13030687E-02 -0.50103778E+01

log likelihood function =  -0.12303208E+04

LR test of the one-sided error =   0.10219192E+05
with number of restrictions = 2
[note that this statistic has a mixed chi-square distribution]

number of iterations =     15

(maximum number of iterations set at :   100)

number of cross-sections =    533

number of time periods =      7

total number of observations =   3731

thus there are:      0  obsns not in the panel


covariance matrix :

  0.32687875E-01 -0.38692395E-02  0.15931380E-02 -0.37544925E-02 -0.31462486E-04
  0.69388376E-04
-0.38692395E-02  0.51047896E-03 -0.31749878E-03  0.52967412E-03  0.43254338E-05
-0.75647287E-05
  0.15931380E-02 -0.31749878E-03  0.44226000E-03 -0.80715245E-04 -0.56201467E-06
  0.97568566E-06
-0.37544925E-02  0.52967412E-03 -0.80715245E-04  0.31714885E-01  0.23785310E-03
-0.36721648E-04
-0.31462486E-04  0.43254338E-05 -0.56201467E-06  0.23785310E-03  0.20680291E-05
-0.27334778E-06
  0.69388376E-04 -0.75647287E-05  0.97568566E-06 -0.36721648E-04 -0.27334778E-06
  0.16979881E-05




数据结果，请教各位大神，请问大家这个结果要从哪些方面分析，这个结果怎么样？