我的GMM模型,个人感觉结果不错,其中的我所关心的解释变量Iov的系数显著为负,sargen检验和hansen检验也基本合格,请斑竹帮我看看,这个GMM模型的工具变量是否合适?也不存在二阶序列相关。
xtabond2 effi L.effi Iov STK CRED GDPP FDI trade gov edu infra urban yr*, gmm(L.effi,col
> lapse lag(2 5)) iv(Iov STK CRED FDI trade GDPP gov edu infra urban yr*) robust small
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, per
> m.
yr1 dropped due to collinearity
yr13 dropped due to collinearity
Warning: Two-step estimated covariance matrix of moments is singular.
Using a generalized inverse to calculate robust weighting matrix for Hansen test.
Difference-in-Sargan statistics may be negative.
Dynamic panel-data estimation, one-step system GMM
------------------------------------------------------------------------------
Group variable: region Number of obs = 337
Time variable : year Number of groups = 29
Number of instruments = 27 Obs per group: min = 8
F(22, 28) = 7.86 avg = 11.62
Prob > F = 0.000 max = 12
------------------------------------------------------------------------------
| Robust
effi | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
effi |
L1. | -.352063 .1982388 -1.78 0.087 -.7581367 .0540107
|
Iov | -.5553496 .1389817 -4.00 0.000 -.8400406 -.2706585
STK | 1.242945 .4271349 2.91 0.007 .3679992 2.117891
CRED | .010677 .0371951 0.29 0.776 -.0655137 .0868676
GDPP | 3.16e-06 6.26e-06 0.50 0.618 -9.67e-06 .000016
FDI | -.7796826 .815882 -0.96 0.347 -2.450941 .8915759
trade | -.2295421 .1867281 -1.23 0.229 -.6120374 .1529531
gov | .7800951 .4966428 1.57 0.127 -.2372316 1.797422
edu | .0218627 .0305915 0.71 0.481 -.0408012 .0845266
infra | -.0116664 .0728245 -0.16 0.874 -.1608407 .1375078
urban | -.1552193 .1578775 -0.98 0.334 -.4786167 .1681781
yr2 | -.1582223 .0930737 -1.70 0.100 -.3488752 .0324306
yr3 | -.1833154 .0813954 -2.25 0.032 -.3500463 -.0165845
yr4 | .0246586 .0800301 0.31 0.760 -.1392757 .1885928
yr5 | .1058062 .094941 1.11 0.275 -.0886717 .300284
yr6 | -.0309077 .0924328 -0.33 0.741 -.2202477 .1584322
yr7 | -.023427 .0903909 -0.26 0.797 -.2085845 .1617304
yr8 | -.0802579 .0899268 -0.89 0.380 -.2644647 .1039489
yr9 | .0742523 .0866163 0.86 0.399 -.1031733 .2516778
yr10 | .000758 .0922563 0.01 0.994 -.1882204 .1897364
yr11 | -.0692701 .0732593 -0.95 0.352 -.219335 .0807949
yr12 | -.0771347 .0535861 -1.44 0.161 -.1869008 .0326315
_cons | .5240451 .2135728 2.45 0.021 .0865611 .961529
------------------------------------------------------------------------------
Instruments for first differences equation
Standard
D.(Iov STK CRED FDI trade GDPP gov edu infra urban yr1 yr2 yr3 yr4 yr5 yr6
yr7 yr8 yr9 yr10 yr11 yr12 yr13)
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(2/5).L.effi collapsed
Instruments for levels equation
Standard
_cons
Iov STK CRED FDI trade GDPP gov edu infra urban yr1 yr2 yr3 yr4 yr5 yr6
yr7 yr8 yr9 yr10 yr11 yr12 yr13
GMM-type (missing=0, separate instruments for each period unless collapsed)
DL.L.effi collapsed
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -1.12 Pr > z = 0.263
Arellano-Bond test for AR(2) in first differences: z = -1.84 Pr > z = 0.066
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(4) = 4.32 Prob > chi2 = 0.364
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(4) = 2.37 Prob > chi2 = 0.669
(Robust, but can be weakened by many instruments.)
Difference-in-Hansen tests of exogeneity of instrument subsets:
GMM instruments for levels
Hansen test excluding group: chi2(3) = 2.36 Prob > chi2 = 0.501
Difference (null H = exogenous): chi2(1) = 0.01 Prob > chi2 = 0.943
.
end of do-file
.