simultaneous-equation model for dynamic panel data

[ermia astin]

Hi Dear members,

Is there a module in Stata that estimates simultaneous-
equation model for dynamic panel data?



Thank you,
Ermia

[Phil Bromiley]

Look at ivregress, xtivregress, xtabond, xtdpdsys - which works for you depends on exactly what you want to do. Alternatively, you might do it in the structural equation package.
Phil

[Annie Farooq]

Hi Phil,
I hope you would help me.
I need help regarding estimation procedure of the following dynamic simultaneous equations model. I have a balanced panel data, N=14 and T=30. I want to use this model to explore causality links. X,Y and Z are endogenous variables, remaining variables are exogenous.

g(X)_it= β_o+β_1ig(X)_it-1+β_2ig(Y)_it+β_3ig(Z)_it+β_4ig(K)_it+ϵ_it
g(Y)_it= ψ_o+ψ₁g(Y)_it-1+ψ_2ig(X)_it+ψ_3ig(Z)_it+ψ_4ig(K)_it+ψ_5i(L)t+ ψ_6i(M) +ϵ_it
g(Z)_it = λo+ λ₁g(Z)_it-1 + λ_2ig(X)_it+ λ_3ig(Y)_it+ λ_4i(N)_it+ λ_5i(O)_it+ϵ_it

can it be estimated using GMM estimator (Arellano Bond)? and how? please explain. or any alternative technique?
P.S i am using Stata 11.
Thanks

[Bruno De Menna]

Dear Statalist forum members,

I would like to know whether the gmm command is relevant to estimate a system of two simulteneous dynamic panel-data equations ?

If so, are there any recommendations or attention points I should pay attention to when using such command (particularly with regard to xtinstruments) ?

Does the combination of two dynamic models in the same simultaneous equations system endanger the quality of the results ?

Thank you very much for your expertise,

Bruno

[Bruno De Menna]

More precisely, here is my coding and results:

gmm (Risk: LLPGrossLoans_w - {Risk: L.LLPGrossLoans_w ROAA_w L.Size L.NetLoansAssets_w L.CosttoIncome_w EURIBOR3M RealGDP}) (Profitability: ROAA_w - {Profitability: L.ROAA_w LLPGrossLoans_w L.EquityAssets_w L.LiquidAssetsAssets_w Spread10YBond_M3Bond RealGDP}), instruments(Risk: L.LLPGrossLoans_w ROAA_w L.Size L.NetLoansAssets_w L.CosttoIncome_w EURIBOR3M RealGDP) instruments(Profitability: L.ROAA_w LLPGrossLoans_w L.EquityAssets_w L.LiquidAssetsAssets_w Spread10YBond_M3Bond RealGDP) winitial(unadjusted, independent) wmatrix(unadjusted) vce(cluster FitchID)


Step 1
Iteration 0: GMM criterion Q(b) = .6777485
Iteration 1: GMM criterion Q(b) = .00379026
Iteration 2: GMM criterion Q(b) = .00379026

Step 2
Iteration 0: GMM criterion Q(b) = .00956234
Iteration 1: GMM criterion Q(b) = .0095608
Iteration 2: GMM criterion Q(b) = .0095608

GMM estimation

Number of parameters = 13
Number of moments = 15
Initial weight matrix: Unadjusted Number of obs = 26,156
GMM weight matrix: Unadjusted

(Std. Err. adjusted for 3,999 clusters in FitchID)
--------------------------------------------------------------------------------------------
| Robust
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
--------------------------+----------------------------------------------------------------
Risk |
LLPGrossLoans_w |
L1. | .5644038 .0103106 54.74 0.000 .5441955 .5846122
|
ROAA_w | -.1343093 .0176579 -7.61 0.000 -.1689182 -.0997004
|
Size |
L1. | .0234721 .0014808 15.85 0.000 .0205698 .0263743
|
NetLoansAssets_w |
L1. | -.000946 .0003101 -3.05 0.002 -.0015537 -.0003383
|
CosttoIncome_w |
L1. | -.0006993 .0003437 -2.03 0.042 -.001373 -.0000256
|
EURIBOR3M | -.022703 .0070673 -3.21 0.001 -.0365547 -.0088513
RealGDP | -.1293503 .0035371 -36.57 0.000 -.1362828 -.1224178
--------------------------+----------------------------------------------------------------
Profitability |
ROAA_w |
L1. | .6968808 .0119817 58.16 0.000 .6733971 .7203644
|
LLPGrossLoans_w | -.0533006 .0042883 -12.43 0.000 -.0617055 -.0448958
|
EquityAssets_w |
L1. | .0082321 .0006117 13.46 0.000 .0070331 .0094311
|
LiquidAssetsAssets_w |
L1. | .0004012 .000178 2.25 0.024 .0000523 .0007501
|
Spread10YBond_M3Bond | .0156652 .0023249 6.74 0.000 .0111084 .020222
RealGDP | .0085486 .0015714 5.44 0.000 .0054687 .0116285
--------------------------------------------------------------------------------------
Instruments for equation Risk: L.LLPGrossLoans_w ROAA_w L.Size L.NetLoansAssets_w
L.CosttoIncome_w EURIBOR3M RealGDP _cons
Instruments for equation Profitability: L.ROAA_w LLPGrossLoans_w L.EquityAssets_w
L.LiquidAssetsAssets_w Spread10YBond_M3Bond RealGDP _cons




Is there anything wrongly coded here ?


Thanks for your support,

Bruno