I have estimated a dynamic linear panel regression with new command "xtdpdgmm" following the slides provided by
PHP Code:
Sebastian Kripfganz
* asymptotically invalid if the one-step weighting matrix is not optimal
Code:
xtdpdgmm L(0/1).Y X1 X2, model(fodev) gmm(X1 X2, lag(0 2)) vce(r)
Code:
Generalized method of moments estimation Fitting full model: Step 1 f(b) = .00862567 Group variable: id Number of obs = 3877 Time variable: year Number of groups = 138 Moment conditions: linear = 277 Obs per group: min = 1 nonlinear = 0 avg = 28.0942 total = 277 max = 47 (Std. Err. adjusted for 138 clusters in id) ------------------------------------------------------------------------------ | Robust Y | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- Y | L1. | .2349892 .055445 4.24 0.000 .1263189 .3436594 | X1| .0199681 .0091513 2.18 0.029 .002032 .0379043 X2| -.0052541 .0016753 -3.14 0.002 -.0085377 -.0019705 _cons | .0029555 .0078291 0.38 0.706 -.0123893 .0183003 ------------------------------------------------------------------------------ Instruments corresponding to the linear moment conditions: 1, omitted for space 2, model(level): _cons . estat serial, ar(1/3) Arellano-Bond test for autocorrelation of the first-differenced residuals H0: no autocorrelation of order 1: z = -1.8644 Prob > |z| = 0.0623 H0: no autocorrelation of order 2: z = 0.6648 Prob > |z| = 0.5062 H0: no autocorrelation of order 3: z = 2.2343 Prob > |z| = 0.0255 . estat overid Sargan-Hansen test of the overidentifying restrictions H0: overidentifying restrictions are valid 1-step moment functions, 1-step weighting matrix chi2(273) = 616.0026 note: * Prob > chi2 = 0.0000 1-step moment functions, 2-step weighting matrix chi2(273) = 130.9023 note: * Prob > chi2 = 1.0000 * asymptotically invalid if the one-step weighting matrix is not optimal
0 Response to Estimations with XTDPDGMM command
Post a Comment