I am working on a system-GMM and the model does not include the lagged dependent variable. So the model is not dynamic. I went through the previous posts about this issue and found out that xtabond2 command is still applicable to use for static models. However, the output clearly suggests that the model suffers from second-order serial correlation. I know this could be overcome by including further lags of dependent variable but I am not sure if it is an appropriate solution to apply in my case. How should I proceed with the significant AR(2)? Is there any other command I can use?
Here is the output:
Code:
xtabond2 wndf winv wdiv wcf wmtb wsize wprof war wol wtang s2-s6 y3-y32, gmm(winv wdiv, lag(2 5) collaps
> e) iv(wmtb wsize wcf wprof war wol wtang s2-s6 y3-y32, eq(level)) iv(wmtb wsize wprof wcf war wol wtang
> , passthru eq(diff)) robust twostep small
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
Dynamic panel-data estimation, two-step system GMM
------------------------------------------------------------------------------
Group variable: companyno Number of obs = 5624
Time variable : datayearfi~l Number of groups = 499
Number of instruments = 60 Obs per group: min = 1
F(44, 498) = 6.75 avg = 11.27
Prob > F = 0.000 max = 31
------------------------------------------------------------------------------
| Corrected
wndf | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
winv | .4376419 .103269 4.24 0.000 .2347452 .6405386
wdiv | -.0477658 .3213661 -0.15 0.882 -.6791663 .5836346
wcf | .0211632 .0328962 0.64 0.520 -.0434691 .0857956
wmtb | -.0085436 .0096806 -0.88 0.378 -.0275634 .0104762
wsize | .0004914 .0073699 0.07 0.947 -.0139885 .0149714
wprof | -.1751627 .0922974 -1.90 0.058 -.3565029 .0061776
war | -.1093765 .0432954 -2.53 0.012 -.1944407 -.0243124
wol | -.0650945 .0223084 -2.92 0.004 -.1089247 -.0212643
wtang | -.1478948 .065725 -2.25 0.025 -.2770273 -.0187623
s2 | -.0243968 .0470204 -0.52 0.604 -.1167797 .067986
s3 | -.0393369 .0453831 -0.87 0.386 -.1285028 .0498291
s4 | -.0525314 .046443 -1.13 0.259 -.1437797 .038717
s5 | -.0485714 .0463337 -1.05 0.295 -.1396051 .0424622
s6 | -.0351004 .0418741 -0.84 0.402 -.1173721 .0471712
y3 | -.1239323 .2230695 -0.56 0.579 -.5622058 .3143411
y4 | -.2066826 .1784896 -1.16 0.247 -.5573681 .1440029
y5 | -.3227292 .189028 -1.71 0.088 -.6941199 .0486616
y6 | -.2483116 .1836581 -1.35 0.177 -.6091519 .1125287
y7 | -.2241975 .1772996 -1.26 0.207 -.572545 .12415
y8 | -.2040736 .1738694 -1.17 0.241 -.5456815 .1375344
y9 | -.2188938 .1756301 -1.25 0.213 -.5639611 .1261735
y10 | -.2163903 .1771234 -1.22 0.222 -.5643915 .1316109
y11 | -.219408 .1717107 -1.28 0.202 -.5567747 .1179586
y12 | -.1423305 .1738751 -0.82 0.413 -.4839497 .1992886
y13 | -.1968947 .179804 -1.10 0.274 -.5501627 .1563733
y14 | -.2050162 .1794417 -1.14 0.254 -.5575722 .1475399
y15 | -.2253615 .1717706 -1.31 0.190 -.562846 .112123
y16 | -.1961101 .1708418 -1.15 0.252 -.5317697 .1395495
y17 | -.2522972 .1740632 -1.45 0.148 -.5942858 .0896915
y18 | -.1930099 .1733364 -1.11 0.266 -.5335706 .1475509
y19 | -.185219 .1719629 -1.08 0.282 -.5230813 .1526432
y20 | -.1883695 .1683058 -1.12 0.264 -.5190465 .1423075
y21 | -.174492 .1676107 -1.04 0.298 -.5038033 .1548194
y22 | -.1721506 .1680627 -1.02 0.306 -.5023499 .1580486
y23 | -.1901924 .1768949 -1.08 0.283 -.5377448 .15736
y24 | -.1573063 .1717939 -0.92 0.360 -.4948364 .1802239
y25 | -.1784788 .1726957 -1.03 0.302 -.5177806 .1608231
y26 | -.2031764 .1739672 -1.17 0.243 -.5449765 .1386237
y27 | -.1855392 .1706326 -1.09 0.277 -.5207877 .1497093
y28 | -.1882344 .17348 -1.09 0.278 -.5290773 .1526086
y29 | -.20306 .1754951 -1.16 0.248 -.547862 .1417421
y30 | -.2185134 .1758265 -1.24 0.215 -.5639665 .1269397
y31 | -.244247 .1723092 -1.42 0.157 -.5827896 .0942957
y32 | -.2281682 .1765915 -1.29 0.197 -.5751244 .118788
_cons | .4011934 .2501861 1.60 0.109 -.0903571 .8927439
------------------------------------------------------------------------------
Instruments for first differences equation
Standard
wmtb wsize wprof wcf war wol wtang
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(2/5).(winv wdiv) collapsed
Instruments for levels equation
Standard
_cons
wmtb wsize wcf wprof war wol wtang s2 s3 s4 s5 s6 y3 y4 y5 y6 y7 y8 y9 y10
y11 y12 y13 y14 y15 y16 y17 y18 y19 y20 y21 y22 y23 y24 y25 y26 y27 y28
y29 y30 y31 y32
GMM-type (missing=0, separate instruments for each period unless collapsed)
DL.(winv wdiv) collapsed
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -4.75 Pr > z = 0.000
Arellano-Bond test for AR(2) in first differences: z = 1.72 Pr > z = 0.086
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(15) = 28.72 Prob > chi2 = 0.017
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(15) = 11.18 Prob > chi2 = 0.740
(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(13) = 10.18 Prob > chi2 = 0.679
Difference (null H = exogenous): chi2(2) = 0.99 Prob > chi2 = 0.608
gmm(winv wdiv, collapse lag(2 5))
Hansen test excluding group: chi2(5) = 3.58 Prob > chi2 = 0.611
Difference (null H = exogenous): chi2(10) = 7.59 Prob > chi2 = 0.669
iv(wmtb wsize wprof wcf war wol wtang, passthru eq(diff))
Hansen test excluding group: chi2(8) = 4.93 Prob > chi2 = 0.765
Difference (null H = exogenous): chi2(7) = 6.25 Prob > chi2 = 0.511
0 Response to Significant AR(2) in System-GMM where the model does not include the lagged DV
Post a Comment