连老师:
您好!
在学习stata高级课程中我有几点疑问,麻烦您给解答一下,谢谢!
1.在xtabond2命令给出的结果中,sargan检验显示拒绝原假设,而hansen检验又接受原假设,我到底相信哪个判断?
2.在模拟把几个估计值放在同一张表显示,我试了几次都没有成功,麻烦您看毛病出在哪?
程序是:
xtreg roa ln_assets dbr se growth age, fe
est store fix1
xtreg roa ln_assets dbr se growth age, re
est store fix2
local mm "fix1 fix2"
esttab "mm", mtitle("mm") compress
最后预期结果应该是fe和re估计应在一张表上显示,但我老得出下面这张半边的表,还有半边怎么都不显示,很困惑!
. esttab "mm", mtitle("mm") compress
(1)
mm
ln_assets 0.0119***
(10.64)
dbr -0.121***
(-24.25)
se 0.0763***
(7.71)
growth 0.00704***
(6.38)
age -0.00249***
(-10.60)
_cons -0.155***
(-6.84)
N 5090
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001
3.我在用xtabond2做一个练习时出现了这样的警告,虽然能翻译,但是不知道怎样进行改进(应该是对警告含义一知半解造成的)
Warning: Two-step estimated covariance matrix of moments is singular.
Using a generalized inverse to calculate optimal weighting matrix for two-step estimation.
Difference-in-Sargan statistics may be negative.
完整的格式如下:
xtabond2 q Lag_q q2 bsize indep ceo_chair ln_assets dbr se age t2 t3 t4 t5 t6 t7 t8 t9 t10, gmm(q
> bsize indep ceo_chair ln_assets dbr se, lag(4 5)) iv(t2 t3 t4 t5 t6 t7 t8 t9 t10 age) twostep robust
> small
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
t2 dropped due to collinearity
t10 dropped due to collinearity
Warning: Two-step estimated covariance matrix of moments is singular.
Using a generalized inverse to calculate optimal weighting matrix for two-step estimation.
Difference-in-Sargan statistics may be negative.
Dynamic panel-data estimation, two-step system GMM
Group variable: stkcd1 Number of obs = 4072
Time variable : accper1 Number of groups = 509
Number of instruments = 128 Obs per group: min = 8
F(16, 508) = 59.22 avg = 8.00
Prob > F = 0.000 max = 8
Corrected
q Coef. Std. Err. t P>t [95% Conf. Interval]
Lag_q .1914581 .0706903 2.71 0.007 .0525768 .3303394
q2 1.542339 .8957714 1.72 0.086 -.2175331 3.302212
bsize .0074568 .016461 0.45 0.651 -.0248832 .0397968
indep 1.079733 .7000968 1.54 0.124 -.2957086 2.455175
ceo_chair -.0282902 .067454 -0.42 0.675 -.1608135 .104233
ln_assets -.085689 .0342625 -2.50 0.013 -.1530026 -.0183754
dbr -.3226556 .1876035 -1.72 0.086 -.6912298 .0459186
se 3.115585 1.069809 2.91 0.004 1.013789 5.21738
age .0150866 .0065077 2.32 0.021 .0023013 .0278718
t3 .95765 .234903 4.08 0.000 .496149 1.419151
t4 .5948121 .1286597 4.62 0.000 .3420416 .8475826
t5 .507998 .1295486 3.92 0.000 .2534811 .7625149
t6 .3552608 .1112178 3.19 0.001 .1367574 .5737643
t7 .2480908 .1100718 2.25 0.025 .0318388 .4643427
t8 .3882641 .096715 4.01 0.000 .1982536 .5782747
t9 .9071952 .0804689 11.27 0.000 .7491023 1.065288
_cons 1.663733 .7429206 2.24 0.026 .2041575 3.123308
Instruments for first differences equation
Standard
D.(t2 t3 t4 t5 t6 t7 t8 t9 t10 age)
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(4/5).(q bsize indep ceo_chair ln_assets dbr se)
Instruments for levels equation
Standard
_cons
t2 t3 t4 t5 t6 t7 t8 t9 t10 age
GMM-type (missing=0, separate instruments for each period unless collapsed)
DL3.(q bsize indep ceo_chair ln_assets dbr se)
Arellano-Bond test for AR(1) in first differences: z = -5.28 Pr > z = 0.000
Arellano-Bond test for AR(2) in first differences: z = -3.77 Pr > z = 0.000
Sargan test of overid. restrictions: chi2(111) = 520.05 Prob > chi2 = 0.000
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(111) = 132.59 Prob > chi2 = 0.080
(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(69) = 107.21 Prob > chi2 = 0.002
Difference (null H = exogenous): chi2(42) = 25.38 Prob > chi2 = 0.980
iv(t2 t3 t4 t5 t6 t7 t8 t9 t10 age)
Hansen test excluding group: chi2(103) = 116.16 Prob > chi2 = 0.177
Difference (null H = exogenous): chi2(8) = 16.43 Prob > chi2 = 0.037


雷达卡





京公网安备 11010802022788号







