Dear experts,
I am using STATA command xtabond2 and system GMM for my very first project. According to Arellano and Bond (1991), Arellano and Bover (1995) and Blundell and Bond (1998), two necessary tests (Sargan/Hansen and AR2) should be used.
Based on my reading, Sargan and Hansen are used to test the overall validity of the instruments.
The null hypothesis is: Instruments as a group are exogenous.
Hence, the insignificant pvalue is preferred. Nevertheless, when I apply them, Sargan test has pvalue ~ zero while Hansen has pvalue > 0.1
Example:
Sargan test of overid. restrictions: chi2(154) = 513.12 Prob > chi2 = 0.000
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(154) = 148.97 Prob > chi2 = 0.599
(Robust, but weakened by many instruments.)
I search some papers that use SYS GMM and most of them show the results of only Hansen test but not Sargan test.
Should I use only Hansen test or I must use two of them (and try to find instruments that have better Sargan pvalue and Hansen pvalue- insignificant)
Thank you very much for your supports.