为了研究不同投入要素产出弹性的变化情况,运用frontier4.1估计一个超越对数生产函数,最终参数的估计结果如下:
the final mle estimates are :
coefficient standard-error t-ratio
beta 0 0.54941836E+01 0.93158942E+00 0.58976449E+01
beta 1 0.62652658E+00 0.20716165E+00 0.30243367E+01
beta 2 0.75407268E+00 0.21779104E+00 0.34623678E+01
beta 3 -0.65318626E+00 0.19481312E+00 -0.33528864E+01
beta 4 0.28702123E+00 0.39189769E-01 0.73238816E+01
beta 5 -0.33947085E-01 0.29877393E-01 -0.11362131E+01
beta 6 -0.13824178E+00 0.35635495E-01 -0.38793281E+01
beta 7 0.56474596E-01 0.13630715E-01 0.41431866E+01
beta 8 -0.53435461E-03 0.89022321E-03 -0.60024790E+00
beta 9 0.19849231E+00 0.50786376E-01 0.39083771E+01
beta10 -0.53004556E-01 0.27918289E-01 -0.18985604E+01
beta11 -0.12332983E-01 0.56439248E-02 -0.21851784E+01
beta12 -0.52605003E-01 0.34135325E-01 -0.15410722E+01
beta13 0.11002745E-01 0.59363418E-02 0.18534554E+01
beta14 -0.85463582E-02 0.41254247E-02 -0.20716311E+01
sigma-squared 0.51400913E+00 0.69290232E-01 0.74182047E+01
gamma 0.94810572E+00 0.73666775E-02 0.12870194E+03
mu 0.13961877E+01 0.15693568E+00 0.88965597E+01
eta -0.16588563E-01 0.48260760E-02 -0.34372776E+01
log likelihood function = 0.10406508E+03
LR test of the one-sided error = 0.84192693E+03
with number of restrictions = 3
[note that this statistic has a mixed chi-square distribution]
number of iterations = 29
(maximum number of iterations set at : 100)
number of cross-sections = 36
number of time periods = 14
total number of observations = 504
thus there are: 0 obsns not in the panel
由于第一次使用这个软件,想请教各位高手以下问题:
1.t-ratio 什么意思?怎样根据它来判断参数的显著性?
2.看到文献中人家的似然函数值都好几百,这里只有0.1多一点,是不是模型有什么问题啊?
3.有没有关于如何解读结果的资料啊,请推荐一下。
论文急用,希望高手们帮帮忙啊!