动态面板xtabond2命令中的gmm如何表达内生性?

jose.liupei 发表于 2013-7-19 19:48
gmmstyle(varlist [, laglimits(# #) collapse equation({diff | level |both}) passthru split])

varli ...

请问passthru和split有什么作用？

流潋 发表于 2013-7-27 21:15
请问passthru和split有什么作用？

The passthru suboption of gmmstyle() is meaningful only in system GMM, and only for variables for which equation(level) has also been specified.  It directs xtabond2 to create instruments for the levels equation that use not the first-differences of the specified variables but the original levels of the same levels be used as instruments.  Under the standard assumptions, these instruments are not valid.

The split suboption of gmmstyle() is also meaningful only in system GMM, and then only when neither eq(diff) nor eq(level) is specified. Its sole effect is to split the specified instrument group in two for purposes of difference-in-Sargan/Hansen testing--one instrument set for the transformed equation and one for the levels equation.

jose.liupei 发表于 2013-7-29 01:51
The passthru suboption of gmmstyle() is meaningful only in system GMM, and only for variables for  ...

谢谢解答，我明白了。

我也碰到这个问题了，请问你的解决了吗。
我的命令是：xtabond2 y L.y x1 x2 x3 x4 x5 x6,gmm(L.y x5) iv(x1 x2 x3 x4 x6) small twostep
stata 可以运行，但是我不知道到这样做对不对，x5我认为是一个内生变量。
还希望您能指点一下。

ahab88651036 发表于 2012-3-12 14:29
将所有内生变量都放在命令中的gmm( )内就可以了，所用的工具变量只能是这些变量的滞后项。

能不能举一个例子，比如写一下语法

yinshixue 发表于 2013-7-31 21:59
能不能举一个例子，比如写一下语法

如果你对内生性的假设都没问题的话，如下命令：
xtabond2 y L.y x1 x2 x3 x4 x5 x6,gmm(y x5, lag(2 2)) iv(x1 x2 x3 x4 x6) small twostep

jose.liupei 发表于 2013-8-1 20:59
如果你对内生性的假设都没问题的话，如下命令：
xtabond2 y L.y x1 x2 x3 x4 x5 x6,gmm(y x5, lag(2 2)) ...

gmm里面的lag(2 2)是什么含义

yinshixue 发表于 2013-8-1 21:21
gmm里面的lag(2 2)是什么含义

用自身的滞后两期值作为工具变量

jose.liupei 发表于 2013-8-2 02:00
用自身的滞后两期值作为工具变量

请问里边的两个2分别表示什么意思？还是不太明白 非常感谢！！

tony2040044 发表于 2013-12-8 18:03
请问里边的两个2分别表示什么意思？还是不太明白 非常感谢！！

The optional laglimits(a b) suboption means lagged levels dated t-a to t-b are used as instruments.

With lag (2 2) I use lagged levels dated t-2 to t-2 as instruments, which means I use only the second lag of the endogenous variables as instruments.

Or you can use more instruments, like lag (2 3), which means you can use the second and the third lag of the endogenous variables as instruments.