6# vapory
这位亲，我昨完xtabond2后发现了一个非常神奇的事情，想请问一下你。你看，我做的命令和结果，主要看Hansen J-test的数值就可以（红色），居然和我用xtabond做完之后用sargan检验得出的检验数值是一样的（依然是红色），这怎么解释？问题是各个变量的系数没有发生变化，只是显著性变化很多，不理解？？亲，帮我看看吧，谢谢。
. xtabond2 logtotal L.logtotal logincome logtaxshare loggrants logpop if meanpia<0.609,gmmstyle(L.logtotal) ivstyle(log
> income logtaxshare loggrants logpop) nolevel robust small
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
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/Hansen statistics may be negative.
Dynamic panel-data estimation, one-step difference GMM
------------------------------------------------------------------------------
Group variable: number Number of obs = 180
Time variable : year Number of groups = 30
Number of instruments = 25 Obs per group: min = 6
F(5, 30) = 5.94 avg = 6.00
Prob > F = 0.001 max = 6
------------------------------------------------------------------------------
| Robust
logtotal | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
logtotal |
L1. | .1901806 .1818343 1.05 0.304 -.1811746 .5615357
logincome | .5071367 .2904991 1.75 0.091 -.0861417 1.100415
logtaxshare | -.0405732 .0359664 -1.13 0.268 -.1140263 .03288
loggrants | .0346025 .1506271 0.23 0.820 -.2730191 .3422241
logpop | -.7217518 .9117566 -0.79 0.435 -2.583807 1.140304
------------------------------------------------------------------------------
Instruments for first differences equation
Standard
D.(logincome logtaxshare loggrants logpop)
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(1/.).L.logtotal
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -1.61 Pr > z = 0.108
Arellano-Bond test for AR(2) in first differences: z = 2.63 Pr > z = 0.009
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(20) = 37.96 Prob > chi2 = 0.009
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(20) = 24.86 Prob > chi2 = 0.207
(Robust, but can be weakened by many instruments.)
Difference-in-Hansen tests of exogeneity of instrument subsets:
（另一个问题，下面这个hansen tests这个检验的原假设是什么？）
iv(logincome logtaxshare loggrants logpop)
Hansen test excluding group: chi2(16) = 19.26 Prob > chi2 = 0.255
Difference (null H = exogenous): chi2(4) = 5.60 Prob > chi2 = 0.231
.
xtabond logtotal logincome logtaxshare loggrants logpop if meanpia<0.609,lags(1) maxldep(99) twostep
Arellano-Bond dynamic panel-data estimation Number of obs = 180
Group variable: number Number of groups = 30
Time variable: year
Obs per group: min = 6
avg = 6
max = 6
Number of instruments = 26 Wald chi2(5) = 674.53
Prob > chi2 = 0.0000
Two-step results
------------------------------------------------------------------------------
logtotal | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
logtotal |
L1. | .1956033 .0137852 14.19 0.000 .1685848 .2226218
logincome | .5233971 .0396806 13.19 0.000 .4456246 .6011696
logtaxshare | -.0355417 .0078562 -4.52 0.000 -.0509396 -.0201437
loggrants | .0554351 .0219997 2.52 0.012 .0123165 .0985536
logpop | -.6220344 .2177142 -2.86 0.004 -1.048746 -.1953224
_cons | 8.32892 2.659303 3.13 0.002 3.116782 13.54106
------------------------------------------------------------------------------
Warning: gmm two-step standard errors are biased; robust standard
errors are recommended.
Instruments for differenced equation
GMM-type: L(2/100).logtotal
Standard: D.logincome D.logtaxshare D.loggrants D.logpop
Instruments for level equation
Standard: _cons
. estat sargan
Sargan test of overidentifying restrictions
H0: overidentifying restrictions are valid
chi2(20) = 24.86401
Prob > chi2 = 0.2067
扫码加好友，拉您进群
全部版块
我的主页
收藏
万分感谢！
