正在写毕业论文，进行GMM估计，由于没有系统学习过GMM所以有很多疑问回归的因变量是经济增长（lnGDP），自变量是国有经济固定投资比重（ginv）以及他的平方项（ginv2）、总投资的增长（lnInvest），控制变量为dTrade（进出口总额的一阶差分），dTech（专利数量一阶差分），dConsume（消费总额一阶差分），加入了区域程度的两个虚拟变量t10*ginv、t20*ginv
这些回归后得到的GMM的AR(2)和sargan不过关，请指教
进行GMM估计得到结果如下：
1 未加入虚拟变量时
xtabond2 lnGDP ginv ginv2 lninv dcon dTrade dPATENT ,gmm( lnGDP ginv ginv2 lninv dcon dTrade ,lag(2 2)) iv( t10ginv t20g
> inv)
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
Warning: Number of instruments may be large relative to number of observations.

Dynamic panel-data estimation, one-step system GMM
------------------------------------------------------------------------------
Group variable: prov                            Number of obs      =       649
Time variable : year                            Number of groups   =        31
Number of instruments = 239                     Obs per group: min =        20
Wald chi2(6)  =  5.84e+06                                      avg =     20.94
Prob > chi2   =     0.000                                      max =        21
------------------------------------------------------------------------------
       lnGDP |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        ginv |   1.447699   .1308415    11.06   0.000     1.191254    1.704144
       ginv2 |  -2.992862   .1240254   -24.13   0.000    -3.235947   -2.749776
       lninv |    .570674   .0064089    89.04   0.000     .5581128    .5832351
        dcon |    .000406    .000012    33.87   0.000     .0003825    .0004294
      dTrade |   2.32e-08   1.81e-09    12.82   0.000     1.97e-08    2.68e-08
     dPATENT |   3.56e-06   5.73e-07     6.21   0.000     2.44e-06    4.68e-06
       _cons |   3.832363   .0685514    55.90   0.000     3.698004    3.966721
------------------------------------------------------------------------------
Instruments for first differences equation
  Standard
    D.(t10ginv t20ginv)
  GMM-type (missing=0, separate instruments for each period unless collapsed)
    L2.(lnGDP ginv ginv2 lninv dcon dTrade)
Instruments for levels equation
  Standard
    t10ginv t20ginv
    _cons
  GMM-type (missing=0, separate instruments for each period unless collapsed)
    DL.(lnGDP ginv ginv2 lninv dcon dTrade)
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z =  -8.93  Pr > z =  0.000
Arellano-Bond test for AR(2) in first differences: z =   1.70  Pr > z =  0.089
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(232)  =5639.16  Prob > chi2 =  0.000
  (Not robust, but not weakened by many instruments.)

Difference-in-Sargan tests of exogeneity of instrument subsets:
  GMM instruments for levels
    Sargan test excluding group:     chi2(114)  =1247.70  Prob > chi2 =  0.000
    Difference (null H = exogenous): chi2(118)  =4391.46  Prob > chi2 =  0.000
  iv(t10ginv t20ginv)
    Sargan test excluding group:     chi2(230)  =5562.93  Prob > chi2 =  0.000
    Difference (null H = exogenous): chi2(2)    =  76.23  Prob > chi2 =  0.000
2 加入虚拟变量时
xtabond2 lnGDP l.lnGDP ginv ginv2 lninv dcon dTrade dPATENT t10ginv t20ginv ,gmm(lnGDP ginv ginv2,lag(2 2)) iv( t10ginv
> t20ginv)
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
Warning: Number of instruments may be large relative to number of observations.

Dynamic panel-data estimation, one-step system GMM
------------------------------------------------------------------------------
Group variable: prov                            Number of obs      =       649
Time variable : year                            Number of groups   =        31
Number of instruments = 123                     Obs per group: min =        20
Wald chi2(9)  =  3.33e+07                                      avg =     20.94
Prob > chi2   =     0.000                                      max =        21
------------------------------------------------------------------------------
       lnGDP |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
       lnGDP |
         L1. |   .8753864   .0071002   123.29   0.000     .8614702    .8893026
             |
        ginv |  -.1160903   .0735522    -1.58   0.114    -.2602499    .0280693
       ginv2 |  -.2709507   .0684881    -3.96   0.000    -.4051849   -.1367166
       lninv |   .0540894   .0050338    10.75   0.000     .0442233    .0639555
        dcon |   .0000158   7.50e-06     2.10   0.036     1.05e-06    .0000305
      dTrade |   1.20e-08   1.06e-09    11.30   0.000     9.93e-09    1.41e-08
     dPATENT |  -3.08e-07   3.39e-07    -0.91   0.364    -9.73e-07    3.57e-07
     t10ginv |   .0665013   .0116319     5.72   0.000     .0437032    .0892994
     t20ginv |  -.0056776    .009064    -0.63   0.531    -.0234428    .0120876
       _cons |    .826554   .0438474    18.85   0.000     .7406148    .9124933
------------------------------------------------------------------------------
Instruments for first differences equation
  Standard
    D.(t10ginv t20ginv)
  GMM-type (missing=0, separate instruments for each period unless collapsed)
    L2.(lnGDP ginv ginv2)
Instruments for levels equation
  Standard
    t10ginv t20ginv
    _cons
  GMM-type (missing=0, separate instruments for each period unless collapsed)
    DL.(lnGDP ginv ginv2)
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z =  -7.47  Pr > z =  0.000
Arellano-Bond test for AR(2) in first differences: z =  -3.25  Pr > z =  0.001
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(113)  = 621.46  Prob > chi2 =  0.000
  (Not robust, but not weakened by many instruments.)

Difference-in-Sargan tests of exogeneity of instrument subsets:
  GMM instruments for levels
    Sargan test excluding group:     chi2(53)   = 200.98  Prob > chi2 =  0.000
    Difference (null H = exogenous): chi2(60)   = 420.48  Prob > chi2 =  0.000
  iv(t10ginv t20ginv)
    Sargan test excluding group:     chi2(111)  = 619.31  Prob > chi2 =  0.000
    Difference (null H = exogenous): chi2(2)    =   2.15  Prob > chi2 =  0.341


.




.