gliete 发表于 2010-1-19 16:49 
因为--在生产函数中效率值为:exp(-Ui),在成本函数中效率值为:exp(Ui) 大家可以在纸上画一下指数函数 ...
我利用FRONTIER4.1软件,利用随机前沿方法计算农业生产的成本效率,C-D成本函数模型和超越对数成本函数模型的LR都未通过显著性检验。请问怎么解决?
原始数据见附件。
请各位大神帮帮忙!!!
非常感谢!!!
运行结果如下:
the final mle estimates are :
coefficient standard-error t-ratio
beta 0 -0.28185995E+01 0.47767986E+00 -0.59006036E+01
beta 1 0.95051845E+00 0.18881022E-01 0.50342531E+02
beta 2 0.25080712E+00 0.11519457E+00 0.21772477E+01
beta 3 0.28450085E+00 0.51607010E-01 0.55128335E+01
beta 4 0.32932559E+00 0.59448049E-01 0.55397208E+01
beta 5 0.14504195E+00 0.39786302E-01 0.36455249E+01
beta 6 0.38726480E-01 0.99552261E-01 0.38900653E+00
sigma-squared 0.11704648E+00 0.77235940E-02 0.15154407E+02
gamma 0.86471533E-05 0.35974274E-02 0.24037047E-02
mu is restricted to be zero
eta is restricted to be zero
log likelihood function = -0.14685133E+03
the likelihood value is less than that obtained
using ols! - try again using different starting values