跪拜:面板数据差分GMM求教!!!

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

各位前辈:
小弟做了一个面板数据差分GMM,用EVIEWS5.1做的,上面没有Sargan检验,只有J统计量,请问怎样知道是否通过过度识别检验?第二,上面也没有序列相关的检验,这是否重要?该怎么做?谢谢!另外,序列相关检验怎么做?是先产生一个残差项,再用view下的correlogram吗?然后再选择level?大家帮我参谋一下啊!
下面是我做的一个结果,请各位前辈不吝赐教!小弟再拜!

Dependent Variable: WATER
Method: Panel Generalized Method of Moments
Transformation: First Differences
Date: 03/03/08   Time: 22:27
Sample (adjusted): 1994 2005
Cross-sections included: 29
Total panel (balanced) observations: 348
White period instrument weighting matrix
White period standard errors & covariance (d.f. corrected)
Instrument list: @DYN(WATER,-2) @LEV(FDISTK(-2))
@LEV(CONSUM(-2)) @LEV(CHYJG(-2)) @LEV(OWN(-2))
@LEV(RKMD) @LEV(TECH(-2))
Variable	Coefficient	Std. Error	t-Statistic	Prob.
WATER(-1)	0.916048	0.091356	10.02727	0
FDISTK	0.005361	0.008721	0.614689	0.5392
CONSUM	0.21498	0.073745	2.91517	0.0038
CHYJG	-0.16836	0.124623	-1.35092	0.1776
OWN	0.163465	0.047264	3.458522	0.0006
RKMD	-0.27417	0.167715	-1.63472	0.103
TECH	-0.0786	0.027848	-2.82256	0.005
	Effects Specification
Cross-section fixed (first differences)
Mean dependent var	-0.00174	S.D. dependent var		0.125241
S.E. of regression	0.157595	Sum squared resid		8.469139
J-statistic	25.59777	Instrument rank		30

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏6 回帖

关键词：差分GMM 面板数据 GMM observations coefficients adjusted 统计

谢谢各位赐教啊

谢谢各位

请求赐教!

高手请指教

Eviews手册里讲了Sargan检验的方法：

Note in particular the results labeled “J-statistic” and “Instrument rank”. Since the reported J-statistic is simply the Sargan statistic (value of the GMM objective function at estimated parameters), and the instrument rank of 38 is greater than the number of estimated coefficients (13), we may use it to construct the Sargan test of over-identifying restrictions. It is worth noting here that the J-statistic reported by a panel equation differs from that reported by an ordinary equation by a factor equal to the number of observations. Under the null
hypothesis that the over-identifying restrictions are valid, the Sargan statistic is distributed
as chi(p-k), where k is the number of estimated coefficients and p is the instrument rank. The p-value of 0.22 in this example may be computed using “scalar pval =@chisq(30.11247, 25)”.

针对楼主的情况，7个系数，Instrument rank =30,

scalar pval = @chisq(25.597777, 23)

再用show pval，得到pval=0.32，一般按照5%的显著性水平，就是过度识别限制是有效的。

+200

[此贴子已经被jimmy_young2005于2008-8-26 10:36:56编辑过]

1# seaside
你好，我也想做面板数据GMM。可是我经过PROC --- MAKE SYSTEM-步骤后，出现对话框。在对话框输入工具变量应该怎么输，才不会按确定后，程序出现@stackinst 呢？因为面板数据GMM不认可@stackinst

非常感谢，十分尤其非常，哈哈

做动态面板还是用stata吧 这个还可以做系统广义距 eviews做比较复杂的面板模型还是不行的