. gmm (vol -{b1}*time- {b2}*topixvol+{b3}*iita_sq -{b4}*iita_rdfix -{b5}*iita_volrdfix-{b0}), instruments(
> time topixvol iita_sq iita_rdfix iita_volrdfix)
Step 1
Iteration 0: GMM criterion Q(b) = .10394759
Iteration 1: GMM criterion Q(b) = 5.867e-26
Iteration 2: GMM criterion Q(b) = 4.609e-35
Step 2
Iteration 0: GMM criterion Q(b) = 8.724e-33
Iteration 1: GMM criterion Q(b) = 2.203e-33
GMM estimation
Number of parameters = 6
Number of moments = 6
Initial weight matrix: Unadjusted Number of obs = 266
GMM weight matrix: Robust
------------------------------------------------------------------------------
| Robust
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
/b1 | .0004453 .0001966 2.26 0.024 .0000598 .0008307
/b2 | 5.949765 .7526101 7.91 0.000 4.474676 7.424854
/b3 | -.0082446 .0175029 -0.47 0.638 -.0425497 .0260605
/b4 | -.1035155 .3600184 -0.29 0.774 -.8091387 .6021076
/b5 | .0798806 2.085217 0.04 0.969 -4.00707 4.166831
/b0 | .1068786 .024231 4.41 0.000 .0593866 .1543705
------------------------------------------------------------------------------
Instruments for equation 1: time topixvol iita_sq iita_rdfix iita_volrdfix _cons
接过如上,这个和直接用OLS的结果没什么区别。但是后来做个hanse检验
. estat overid
Test of overidentifying restriction:
Hansen's J chi2(0) = 5.9e-31 (p = .)
成了这样的,是不是有什么问题。
OLS的R-squared 有0.48


雷达卡




京公网安备 11010802022788号







