Hello and Greetings from India!
For my current study on analysis of market share of insurance companies, using the one step system GMM, I have made use of xtabond2 command and the output is appended below. Kindly help me to interpret the following:
a) Whether the postestimation AR1 AR2 is OK, what is the null hypothesis for the same
b) How to interpret the Sargan and Hansen tests here, I think my instruments are collpased properly so whether the post estimation is in line with it
c) How to interpret the - Difference-in-Hansen tests of exogeneity of instrument subsets
The model output is as follows:
xtabond2 L(0/2)l_mktshr l_reinsuti l_roa l_diversif l_finlvg l_liq, gmm (l_reinsuti l_roa l_diversif, lag (2 5) collapse) iv (l_finlvg, eq(diff)) iv (l_liq, eq (level)) robust
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
Dynamic panel-data estimation, one-step system GMM
------------------------------------------------------------------------------
Group variable: compid Number of obs = 236
Time variable : year Number of groups = 24
Number of instruments = 18 Obs per group: min = 1
Wald chi2(7) = 2066.17 avg = 9.83
Prob > chi2 = 0.000 max = 14
------------------------------------------------------------------------------
| Robust
l_mktshr | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
l_mktshr |
L1. | .9721295 .0986449 9.85 0.000 .778789 1.16547
L2. | -.1141199 .0459984 -2.48 0.013 -.2042751 -.0239647
|
l_reinsuti | .2576628 .1368833 1.88 0.060 -.0106235 .5259491
l_roa | .2412237 .3302044 0.73 0.465 -.4059651 .8884125
l_diversif | -.0068372 .1239378 -0.06 0.956 -.2497508 .2360765
l_finlvg | -.1238919 .0537652 -2.30 0.021 -.2292697 -.0185141
l_liq | .0212269 .0618168 0.34 0.731 -.0999318 .1423855
_cons | -1.485467 1.475171 -1.01 0.314 -4.376748 1.405814
------------------------------------------------------------------------------
Instruments for first differences equation
Standard
D.l_finlvg
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(2/5).(l_reinsuti l_roa l_diversif) collapsed
Instruments for levels equation
Standard
l_liq
_cons
GMM-type (missing=0, separate instruments for each period unless collapsed)
DL.(l_reinsuti l_roa l_diversif) collapsed
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -2.62 Pr > z = 0.009
Arellano-Bond test for AR(2) in first differences: z = 0.51 Pr > z = 0.612
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(10) = 21.44 Prob > chi2 = 0.018
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(10) = 12.79 Prob > chi2 = 0.236
(Robust, but weakened by many instruments.)
Difference-in-Hansen tests of exogeneity of instrument subsets:
GMM instruments for levels
Hansen test excluding group: chi2(7) = 7.49 Prob > chi2 = 0.380
Difference (null H = exogenous): chi2(3) = 5.29 Prob > chi2 = 0.151
iv(l_finlvg, eq(diff))
Hansen test excluding group: chi2(9) = 12.63 Prob > chi2 = 0.180
Difference (null H = exogenous): chi2(1) = 0.16 Prob > chi2 = 0.689
iv(l_liq, eq(level))
Hansen test excluding group: chi2(9) = 6.94 Prob > chi2 = 0.644
Difference (null H = exogenous): chi2(1) = 5.85 Prob > chi2 = 0.016
For my current study on analysis of market share of insurance companies, using the one step system GMM, I have made use of xtabond2 command and the output is appended below. Kindly help me to interpret the following:
a) Whether the postestimation AR1 AR2 is OK, what is the null hypothesis for the same
b) How to interpret the Sargan and Hansen tests here, I think my instruments are collpased properly so whether the post estimation is in line with it
c) How to interpret the - Difference-in-Hansen tests of exogeneity of instrument subsets
The model output is as follows:
xtabond2 L(0/2)l_mktshr l_reinsuti l_roa l_diversif l_finlvg l_liq, gmm (l_reinsuti l_roa l_diversif, lag (2 5) collapse) iv (l_finlvg, eq(diff)) iv (l_liq, eq (level)) robust
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
Dynamic panel-data estimation, one-step system GMM
------------------------------------------------------------------------------
Group variable: compid Number of obs = 236
Time variable : year Number of groups = 24
Number of instruments = 18 Obs per group: min = 1
Wald chi2(7) = 2066.17 avg = 9.83
Prob > chi2 = 0.000 max = 14
------------------------------------------------------------------------------
| Robust
l_mktshr | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
l_mktshr |
L1. | .9721295 .0986449 9.85 0.000 .778789 1.16547
L2. | -.1141199 .0459984 -2.48 0.013 -.2042751 -.0239647
|
l_reinsuti | .2576628 .1368833 1.88 0.060 -.0106235 .5259491
l_roa | .2412237 .3302044 0.73 0.465 -.4059651 .8884125
l_diversif | -.0068372 .1239378 -0.06 0.956 -.2497508 .2360765
l_finlvg | -.1238919 .0537652 -2.30 0.021 -.2292697 -.0185141
l_liq | .0212269 .0618168 0.34 0.731 -.0999318 .1423855
_cons | -1.485467 1.475171 -1.01 0.314 -4.376748 1.405814
------------------------------------------------------------------------------
Instruments for first differences equation
Standard
D.l_finlvg
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(2/5).(l_reinsuti l_roa l_diversif) collapsed
Instruments for levels equation
Standard
l_liq
_cons
GMM-type (missing=0, separate instruments for each period unless collapsed)
DL.(l_reinsuti l_roa l_diversif) collapsed
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -2.62 Pr > z = 0.009
Arellano-Bond test for AR(2) in first differences: z = 0.51 Pr > z = 0.612
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(10) = 21.44 Prob > chi2 = 0.018
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(10) = 12.79 Prob > chi2 = 0.236
(Robust, but weakened by many instruments.)
Difference-in-Hansen tests of exogeneity of instrument subsets:
GMM instruments for levels
Hansen test excluding group: chi2(7) = 7.49 Prob > chi2 = 0.380
Difference (null H = exogenous): chi2(3) = 5.29 Prob > chi2 = 0.151
iv(l_finlvg, eq(diff))
Hansen test excluding group: chi2(9) = 12.63 Prob > chi2 = 0.180
Difference (null H = exogenous): chi2(1) = 0.16 Prob > chi2 = 0.689
iv(l_liq, eq(level))
Hansen test excluding group: chi2(9) = 6.94 Prob > chi2 = 0.644
Difference (null H = exogenous): chi2(1) = 5.85 Prob > chi2 = 0.016
