关于这个ivreg2工具变量检测的结果如何解读呢？应该看哪几个关键的数值并如何作解读？

suiyuehong 发表于 2015-8-3 18:21
这个一般用于一个被iv对一个iv，工具变量超过了内生变量的个数，就用楼主的讨论了。

好的，谢谢您的解答~

楼主，你的结果不理想，首先回归系数不显著，其次工具变量选择有问题存在弱工具变量，过度识别也不是太理想。

The Sargan-Hansen test is a test of overidentifying restrictions.  The joint
null hypothesis is that the instruments are valid instruments, i.e.,
uncorrelated with the error term,

The underidentification test is an LM test of whether the
equation is identified, i.e., that the excluded instruments are "relevant",
meaning correlated with the endogenous regressors.  The test is essentially the
test of the rank of a matrix:  under the null hypothesis that the equation is
underidentified

When errors are assumed to be i.i.d., ivreg2 automatically reports an LM version of the Anderson (1951) canonical correlations test. ivreg2 also reports the closely-related Cragg-Donald (1993) Wald test statistic. a result of rejection of the null should be treated with caution,
because weak instrument problems may still be present.  

When the i.i.d. assumption is dropped，ivreg2 reports the LM and Wald versions of the Kleibergen-Paap (2006) rk statistic,

crystal8832 发表于 2014-6-28 23:16
首先要说明：你的估计结果并不理想，参数大部分都不显著。
Underidentification test 即识别不足检验 其原 ...

貌似这里的才是真理啊https://bbs.pinggu.org/thread-3679411-1-1.html

trueeconlover 发表于 2014-6-29 08:24
楼上回答不错 不过有些可以补充一下：

你用了cluster 所以error term应该不是iid那么一般的weak identif ...

请问如果用了cluster是不是要根据Kleibergen-Paap rk Wald F statistic来判断工具变量是否存在弱工具变量问题？如果是的话，请问rk wald要大于几才不存在弱工具变量的问题？麻烦您了，谢谢

1029812370 发表于 2015-7-19 17:37
请问，Cragg-Donald Wald F statistic、Kleibergen-Paap rk Wald F statistic这两个数有判断的经验值吗？ ...

请问您现在弄懂这个了吗？

bluce-lee 发表于 2016-5-11 10:53
When errors are assumed to be i.i.d., ivreg2 automatically reports an LM version of the Anderson (19 ...

请问怎么根据Kleibergen-Paap rk statistic判断弱工具变量问题，谢谢！

葫芦娃大王 发表于 2016-11-8 20:45
请问怎么根据Kleibergen-Paap rk statistic判断弱工具变量问题，谢谢！

当你假定残差项不满足独立同分布时，就看这个来判断是否弱工具，原假设是弱工具，所以拒绝原假设就可以