求助，做SFA时用超越对数生产函数，为何STATA和frontier4.1估计结果差异非常大？

fysh1113 发表于 2009-11-10 09:29
跟你介绍一下我的检查过程，顺便帮你分析问题。

我首先检查你的设定，发现没有问题。核心三点：（1）Bat ...

建议用frontier4.1，我也遇到同样问题，stata几乎没有一个变量显著，而frontier大多数变量显著。

frontier4.1运算结果标准差全部为1或全部接近1的情况我也遇到过，应该不是frontier运算的问题，有可能是数据共线性或函数设定偏差太大。印象中以前我的处理方法是消除共线性还是改变函数设定，记不得了，不过经过调整，再也没有出现标准差全为1的情况了。

lswgdhy 发表于 2011-8-17 10:41
建议用frontier4.1，我也遇到同样问题，stata几乎没有一个变量显著，而frontier大多数变量显著。

fron ...

That's possible.
Thanks for your comments!

fysh1113 发表于 2011-8-18 21:53
That's possible.
Thanks for your comments!

请问frontier4.1可以处理unbalanced panel吗？我怎么感觉它和deap一样，数据一多就挂了~

顶。学习。

mark！ BC1995模型stata做不了，期待工程师解决，frontier的扩展性比stata而言太差了，纠结中~~~

xhw88184004 发表于 2009-11-9 18:56
相关数据附上了，其中：logY,logK,logL分别是总产出，资本存量，劳动投入量（均取对数）；h,insti,gov, ...

你设置错了，应该用模型2也就是说TE EFFECTS MODEL 然后把ETA设置成影响无效率函数的自变量的个数

弱弱问一句TE EFFECTS MODEL 是什么模型？还有怎么用frontier4.0来分解TFP呢？谢谢各位高人啦

学习下~

lxfkxkr 发表于 2012-9-27 11:17
mark！ BC1995模型stata做不了，期待工程师解决，frontier的扩展性比stata而言太差了，纠结中~~~

现在可以做了，请看下面两个更新的程序
sfcross extends the capabilities of the frontier command by including additional models (Greene, 2003, Journal of Productivity Analysis 19: 179–190; Wang, 2002, Journal of Productivity Analysis 18: 241–253) and command functionality, such as the possibility of managing complex survey data characteristics. Similarly, sfpanel allows one to fit a much wider range of time-varying inefficiency models compared with the xtfrontier command, including the model of Cornwell, Schmidt, and Sickles (1990, Journal of Econometrics 46: 185–200); the model of Lee and Schmidt (1993, in The Measurement of Productive Efficiency: Techniques and Applications), a production frontier model with flexible temporal variation in technical efficiency; the flexible model of Kumbhakar (1990, Journal of Econometrics 46: 201–211); the inefficiency effects model of Battese and Coelli (1995 Empirical Economics 20: 325–332); and the "true" fixed- and random-effects models of Greene (2005a, Journal of Econometrics 126: 269–303). A brief overview of the stochastic frontier literature, a description of the two commands and their options, and examples using simulated and real data are provided.