我在仿照网上的gmm检验步骤,在stata中得到了验证结果,stata菜鸟,水平有限,我认为应该是验证成功了,但不敢确定,请各位大神帮忙指点一下代码:
ivregress gmm lnX lngdpi lnpopi lnpopj lnD F L T lnHC lnLTIPR (lnIPR = lnLT lngdpj),robust
estat firststage, all
estat overid
estat endogenous lnIPR
ivreg2 lnX lngdpi lngdpj lnpopi lnpopj lnD F L T lnHC lnLTIPR (lnIPR = lnLT lngdpj),gmm2s
结果:
先检验内生变量
. ivregress gmm lnX lngdpi lnpopi lnpopj lnD F L T lnHC lnLTIPR (lnIPR = lnLT lngdpj),robust
Instrumental variables (GMM) regression Number of obs = 680
Wald chi2(10) = 1561.88
Prob > chi2 = 0.0000
R-squared = 0.6431
GMM weight matrix: Robust Root MSE = 1.1052
------------------------------------------------------------------------------
| Robust
lnX | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lnIPR | 28.16103 4.078869 6.90 0.000 20.16659 36.15546
lngdpi | 10.01916 1.569068 6.39 0.000 6.943843 13.09448
lnpopi | -11.23935 4.260464 -2.64 0.008 -19.58971 -2.888997
lnpopj | .981998 .0314548 31.22 0.000 .9203477 1.043648
lnD | -.3356042 .0879735 -3.81 0.000 -.508029 -.1631794
F | -.5519003 .1996394 -2.76 0.006 -.9431863 -.1606142
L | 2.293873 .2587026 8.87 0.000 1.786825 2.80092
T | -.0570795 .0091794 -6.22 0.000 -.0750707 -.0390883
lnHC | -2.838792 2.740194 -1.04 0.300 -8.209475 2.53189
lnLTIPR | -2.079046 .326958 -6.36 0.000 -2.719872 -1.43822
_cons | -42.79774 64.73064 -0.66 0.509 -169.6675 84.07199
------------------------------------------------------------------------------
Instrumented: lnIPR
Instruments: lngdpi lnpopi lnpopj lnD F L T lnHC lnLTIPR lnLT lngdpj
弱工具性检验
. estat firststage, all
First-stage regression summary statistics
--------------------------------------------------------------------------
| Adjusted Partial Robust
Variable | R-sq. R-sq. R-sq. F(2,668) Prob > F
-------------+------------------------------------------------------------
lnIPR | 0.9944 0.9944 0.0760 28.6272 0.0000
--------------------------------------------------------------------------
Shea's partial R-squared
--------------------------------------------------
| Shea's Shea's
Variable | Partial R-sq. Adj. Partial R-sq.
-------------+------------------------------------
lnIPR | 0.0760 0.0621
--------------------------------------------------
过度识别检验
. estat overid
Test of overidentifying restriction:
Hansen's J chi2(1) = 113.858 (p = 0.0000)
. estat endogenous lnIPR
Test of endogeneity (orthogonality conditions)
Ho: variables are exogenous
GMM C statistic chi2(1) = 22.9764 (p = 0.0000)
GMM验证
.
. ivreg2 lnX lngdpi lngdpj lnpopi lnpopj lnD F L T lnHC lnLTIPR (lnIPR = lnLT lngdpj),gmm2s
Warning - duplicate variables detected
Duplicates: lngdpj
2-Step GMM estimation
---------------------
Estimates efficient for homoskedasticity only
Statistics consistent for homoskedasticity only
Number of obs = 680
F( 11, 668) = 271.18
Prob > F = 0.0000
Total (centered) SS = 2327.342805 Centered R2 = 0.8172
Total (uncentered) SS = 300051.2976 Uncentered R2 = 0.9986
Residual SS = 425.3903337 Root MSE = .7909
------------------------------------------------------------------------------
lnX | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lnIPR | 1.187729 4.99029 0.24 0.812 -8.59306 10.96852
lngdpi | 5.699809 1.352614 4.21 0.000 3.048735 8.350883
lngdpj | .7946897 .0487147 16.31 0.000 .6992107 .8901688
lnpopi | -69.7985 28.34201 -2.46 0.014 -125.3478 -14.24918
lnpopj | .2060715 .050728 4.06 0.000 .1066464 .3054966
lnD | -.5117757 .0585376 -8.74 0.000 -.6265072 -.3970441
F | .3414293 .1306623 2.61 0.009 .0853359 .5975227
L | 2.652754 .1463557 18.13 0.000 2.365902 2.939606
T | -.0146569 .0082155 -1.78 0.074 -.0307589 .0014452
lnHC | -4.400852 2.093598 -2.10 0.036 -8.504228 -.2974758
lnLTIPR | -.0638482 .3973245 -0.16 0.872 -.84259 .7148936
_cons | 1311.531 580.2193 2.26 0.024 174.322 2448.74
------------------------------------------------------------------------------
Underidentification test (Anderson canon. corr. LM statistic): 43.571
Chi-sq(1) P-val = 0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 45.733
Stock-Yogo weak ID test critical values: 10% maximal IV size 16.38
15% maximal IV size 8.96
20% maximal IV size 6.66
25% maximal IV size 5.53
Source: Stock-Yogo (2005). Reproduced by permission.
------------------------------------------------------------------------------
Sargan statistic (overidentification test of all instruments): 0.000
(equation exactly identified)
------------------------------------------------------------------------------
Instrumented: lnIPR
Included instruments: lngdpi lngdpj lnpopi lnpopj lnD F L T lnHC lnLTIPR
Excluded instruments: lnLT
Duplicates: lngdpj
------------------------------------------------------------------------------
.
我觉得应该算验证成功了吧?请各位大神帮忙看看,谢谢了