hello Stata Community,
I am using panel data to estimate one-step and two-step GMM models to study the behavior of Italian banks' NPL (non-performing loans) as a % Gross Customer Loans (GCL). I am using a set of variables (macro and bank-level predictors).
I used the xtabond2 command, but since this is my first time using this command, I would like someone to help me interpret the output. If you are familiar with the xtabond2 command please give me a little help.
This is Stata output
My questions are :
Ar(1) and Ar (2) tell me that there is no second-order serial correlation -if my judgments are correct- which is something I would like to have.
2) I think I have too many instruments and Hensen test tells me there's something wrong in the model. What do you think? How can I reduce the number of instruments?
3) What does the Sargan test tell me?
Please feel free to add or highlight any further points of concern I probably skipped.
Thanks in advance for your help.
Regards,
I am using panel data to estimate one-step and two-step GMM models to study the behavior of Italian banks' NPL (non-performing loans) as a % Gross Customer Loans (GCL). I am using a set of variables (macro and bank-level predictors).
I used the xtabond2 command, but since this is my first time using this command, I would like someone to help me interpret the output. If you are familiar with the xtabond2 command please give me a little help.
This is Stata output
Code:
xtabond2 NPL_perc L1_NPL_perc NetInterestMargin AvgEquityAvgAssets CosttoIncome ROAA L1_NetInterestMargin L1_AvgEquityAvgAssets L1_Cos
> ttoIncome L1_ROAA deltabankloans deltaFTSEMIB RealGDPGrowth deltaNCLDeposits L1_RealGDPGrowth L2_RealGDPGrowth L1_deltabankloans L1_de
> ltaNCLDeposits, twostep ivstyle( RealGDPGrowth deltaFTSEMIB) gmmstyle( AvgEquityAvgAssets NetInterestMargin ROAA CosttoIncome deltaban
> kloans) robust small
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.
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/Hansen statistics may be negative.
Dynamic panel-data estimation, two-step system GMM
------------------------------------------------------------------------------
Group variable: id Number of obs = 1197
Time variable : year Number of groups = 109
Number of instruments = 272 Obs per group: min = 9
F(17, 108) = 1034.59 avg = 10.98
Prob > F = 0.000 max = 11
---------------------------------------------------------------------------------------
| Corrected
NPL_perc | Coefficient std. err. t P>|t| [95% conf. interval]
----------------------+----------------------------------------------------------------
L1_NPL_perc | .7271356 .0539107 13.49 0.000 .6202752 .833996
NetInterestMargin | .0159377 .0062341 2.56 0.012 .0035806 .0282947
AvgEquityAvgAssets | -.0031431 .0026203 -1.20 0.233 -.0083371 .0020509
CosttoIncome | -.000669 .0002176 -3.07 0.003 -.0011003 -.0002377
ROAA | -.0236108 .0059823 -3.95 0.000 -.0354687 -.0117528
L1_NetInterestMargin | -.0021123 .0059489 -0.36 0.723 -.013904 .0096794
L1_AvgEquityAvgAssets | .0021098 .0020469 1.03 0.305 -.0019474 .006167
L1_CosttoIncome | -.0001809 .0002403 -0.75 0.453 -.0006573 .0002955
L1_ROAA | -.0135676 .0066369 -2.04 0.043 -.0267232 -.0004121
deltabankloans | -.0014941 .0007679 -1.95 0.054 -.0030162 .0000281
deltaFTSEMIB | .0001526 .0001276 1.20 0.234 -.0001003 .0004056
RealGDPGrowth | .0027587 .000687 4.02 0.000 .001397 .0041204
deltaNCLDeposits | -.0060776 .0006296 -9.65 0.000 -.0073255 -.0048296
L1_RealGDPGrowth | .0055486 .0013905 3.99 0.000 .0027923 .0083048
L2_RealGDPGrowth | -.0015084 .0009908 -1.52 0.131 -.0034723 .0004554
L1_deltabankloans | -.0002839 .0009126 -0.31 0.756 -.0020928 .0015249
L1_deltaNCLDeposits | -.0012947 .0004159 -3.11 0.002 -.0021192 -.0004702
_cons | .0631293 .0240864 2.62 0.010 .0153858 .1108728
---------------------------------------------------------------------------------------
Instruments for first differences equation
Standard
D.(RealGDPGrowth deltaFTSEMIB)
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(1/10).(AvgEquityAvgAssets NetInterestMargin ROAA CosttoIncome
deltabankloans)
Instruments for levels equation
Standard
RealGDPGrowth deltaFTSEMIB
_cons
GMM-type (missing=0, separate instruments for each period unless collapsed)
D.(AvgEquityAvgAssets NetInterestMargin ROAA CosttoIncome deltabankloans)
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -1.97 Pr > z = 0.049
Arellano-Bond test for AR(2) in first differences: z = 0.63 Pr > z = 0.531
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(254) = 473.29 Prob > chi2 = 0.000
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(254) = 102.47 Prob > chi2 = 1.000
(Robust, but weakened by many instruments.)
Difference-in-Hansen tests of exogeneity of instrument subsets:
GMM instruments for levels
Hansen test excluding group: chi2(211) = 96.70 Prob > chi2 = 1.000
Difference (null H = exogenous): chi2(43) = 5.77 Prob > chi2 = 1.000
iv(RealGDPGrowth deltaFTSEMIB)
Hansen test excluding group: chi2(252) = 101.87 Prob > chi2 = 1.000
Difference (null H = exogenous): chi2(2) = 0.60 Prob > chi2 = 0.740
.
Ar(1) and Ar (2) tell me that there is no second-order serial correlation -if my judgments are correct- which is something I would like to have.
2) I think I have too many instruments and Hensen test tells me there's something wrong in the model. What do you think? How can I reduce the number of instruments?
3) What does the Sargan test tell me?
Please feel free to add or highlight any further points of concern I probably skipped.
Thanks in advance for your help.
Regards,
Comment