cyy1985218 发表于 2011-7-20 13:31 
xtreg tl size ndts tang tobin npr, fe
est store fe
xtivreg tl size ndts tang (npr=tobin L1.npr), fe
est store fe_iv
hausman fe fe_iv
这个检验是P值为0,认为存在内生性,
这两种检验结果不同,到底应该用哪种检验呀
为什么在xtreg那边有tobin,而在xtivreg那边tobin被视为工具变数?
根据stata的帮助文件里提到
Davidson and MacKinnon (1993)
demonstrate that this test, which is similar to the (Durbin-Wu-)Hausman test
(see hausman) in this context, will always yield a computable test statistic,
whereas the Hausman test, depending on the difference of estimated covariance
matrices being a positive definite matrix, often cannot be computed by standard
matrix inverse methods.
因此除非是一些状况下,否则两者应当一样,
我个人强烈怀疑,您在执行Hausman test那边变数摆法似乎有问题…