Dear Statalisters,

I have one doubt about the result of my Arellano-Bond test for autocorrelation in a gmm estimation.
The Hansen test of joint validity is ok with a p-value bigger then 10% but not too high.
The number of instruments is also ok, I have N=95 & T=20 and around 20 instruments.
However the AR(2) has been giving me p-values of order 0.7 to 0.8. Until I know de AR test is not weaken as Hansen, so i don't know if these p-values are reliable or not.

. xtabond2 empshare L.empshare ln_gdppc ln_gdppc_2 ln_pop ln_pop_2 i.year2,
> gmmstyle (L.empshare, lag(1 5)collapse)
> iv(ln_gdppc ln_gdppc_2 ln_pop ln_pop_2 i.year2) orthogonal twostep robust

Arellano-Bond test for AR(1) in first differences: z = -3.18 Pr > z = 0.001
Arellano-Bond test for AR(2) in first differences: z = -0.26 Pr > z = 0.791
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(4) = 14.17 Prob > chi2 = 0.007
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(4) = 6.33 Prob > chi2 = 0.176
(Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:
GMM instruments for levels
Hansen test excluding group: chi2(3) = 5.11 Prob > chi2 = 0.164
Difference (null H = exogenous): chi2(1) = 1.22 Prob > chi2 = 0.269