做三阶段DEA模型,第二阶段随机前沿分析结果如下,请问如何得到投入调整项,做第三阶段。看到有人说冗余项问题,请问从结果中怎么看出冗余项,或者是实际值和最优值。顺便看看结果有没有不合理的地方。求高手指点~~谢谢!
Frontier结果如下:
Output from the program FRONTIER (Version 4.1c)
instruction file = 2.ins
data file = 2.dta
Error Components Frontier (see B&C 1992)
The model is a production function
The dependent variable is not logged
the ols estimates are :
coefficient standard-error t-ratio
beta 0 -0.63646150E+02 0.24816271E+02 -0.25646944E+01
beta 1 0.97407945E-05 0.69556802E-05 0.14004086E+01
beta 2 0.21305168E+03 0.46685713E+02 0.45635306E+01
sigma-squared 0.41304394E+04
log likelihood function = -0.80787543E+03
the estimates after the grid search were :
beta 0 0.81373526E+01
beta 1 0.97407945E-05
beta 2 0.21305168E+03
sigma-squared 0.91978533E+04
gamma 0.88000000E+00
mu is restricted to be zero
eta 0.00000000E+00
the final mle estimates are :
coefficient standard-error t-ratio
beta 0 0.19968629E+03 0.25537022E+02 0.78194823E+01
beta 1 0.12242235E-04 0.39160415E-05 0.31261760E+01
beta 2 0.47394116E+02 0.25347647E+02 0.18697639E+01
sigma-squared 0.15564183E+05 0.14685054E+04 0.10598656E+02
gamma 0.96873619E+00 0.57490231E-02 0.16850449E+03
mu is restricted to be zero
eta 0.16119161E-01 0.69924618E-02 0.23052198E+01
log likelihood function = -0.73780537E+03
LR test of the one-sided error = 0.14014013E+03
with number of restrictions = 2
[note that this statistic has a mixed chi-square distribution]
number of iterations = 100
(maximum number of iterations set at : 100)
number of cross-sections = 29
number of time periods = 5
total number of observations = 145
thus there are: 0 obsns not in the panel