Hi, I am running a sys-gmm regression by using xtabond2.
I have plenty of independent variables yet no literature to identify the endogeneity. How to Do xtabond2 (Roodman, 2009) suggested that Hansen test should between 0.1 and 0.25. Roodman also suggests that put every regressor into the instrument matrix in some form. I have no literature support of choosing endogenous variables, therefore by trial and error, I include the ndepended variable with two lags into gmmstyle option, it gives ok results.
should I report the results (as it looks fine, but i didnt put every regressor into the instrument matrix).
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -0.24 Pr > z = 0.813
Arellano-Bond test for AR(2) in first differences: z = 0.94 Pr > z = 0.346
Arellano-Bond test for AR(3) in first differences: z = 0.92 Pr > z = 0.357
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(9) = 9.20 Prob > chi2 = 0.419
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(9) = 11.45 Prob > chi2 = 0.246
(Robust, but weakened by many instruments.)
I have plenty of independent variables yet no literature to identify the endogeneity. How to Do xtabond2 (Roodman, 2009) suggested that Hansen test should between 0.1 and 0.25. Roodman also suggests that put every regressor into the instrument matrix in some form. I have no literature support of choosing endogenous variables, therefore by trial and error, I include the ndepended variable with two lags into gmmstyle option, it gives ok results.
should I report the results (as it looks fine, but i didnt put every regressor into the instrument matrix).
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -0.24 Pr > z = 0.813
Arellano-Bond test for AR(2) in first differences: z = 0.94 Pr > z = 0.346
Arellano-Bond test for AR(3) in first differences: z = 0.92 Pr > z = 0.357
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(9) = 9.20 Prob > chi2 = 0.419
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(9) = 11.45 Prob > chi2 = 0.246
(Robust, but weakened by many instruments.)
