RE: st: GMM criterion function

["Ornaghi C." <C.Ornaghi@soton.ac.uk>]

I have a problem with the value of the criterion function in GMM.
In the results below you can see that the value of the GMM criterion in the last iteration Q(b) =  1.416e-06 is different from the result saved in e(Q) =  .000210635363449 .
I would appreciate if you could help me clarifying this difference.


Step 1
Iteration 0:   GMM criterion Q(b) =  4.990e-06  
Iteration 1:   GMM criterion Q(b) =  1.455e-06  
Iteration 2:   GMM criterion Q(b) =  1.416e-06  

GMM estimation 

Number of parameters =   5
Number of moments    =   7
Initial weight matrix: Unadjusted                     Number of obs  =   20099

------------------------------------------------------------------------------
             |               Robust
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         /b1 |   .6298857   29.42284     0.02   0.983    -57.03783     58.2976
         /b2 |  -.5015724   .7868027    -0.64   0.524    -2.043677    1.040533
         /b3 |     .58346   78.29432     0.01   0.994    -152.8706    154.0375
         /b4 |   -.500558   19.35732    -0.03   0.979    -38.44021     37.4391
         /b5 |   .9685305   .0525856    18.42   0.000     .8654646    1.071596
------------------------------------------------------------------------------
Instruments for equation 1: l_inp1 l_inp2 l_inp3 l_lninp1 l_lninp2 l_lninp3 _cons

scalars:
               e(rank) =  5
                  e(N) =  20099
                  e(Q) =  .000210635363449
                  e(J) =  4.233560169961503
               e(J_df) =  2
          e(converged) =  1
         e(has_xtinst) =  0
               e(type) =  2
               e(n_eq) =  1
                  e(k) =  5
          e(n_moments) =  7
              e(k_aux) =  5
         e(k_eq_model) =  0
               e(k_eq) =  5

macros:
            e(cmdline) : "gmm gmm_step2, nequation(1) nparameters(5) instrument(l_inp1 l_inp2 l_inp3 l_lninp1 l.."
                e(cmd) : "gmm"
          e(estat_cmd) : "gmm_estat"
            e(predict) : "gmm_p"
       e(marginsnotok) : "_ALL"
            e(eqnames) : "1"
          e(technique) : "dfp"
              e(winit) : "Unadjusted"
          e(estimator) : "onestep"
            e(wmatrix) : "robust"
                e(vce) : "robust"
            e(vcetype) : "Robust"
           e(evalopts) : "depvar(lny2) phi(phihat) phi_lag(phihat_lag)"
           e(evalprog) : "gmm_step2"
             e(params) : "b1 b2 b3 b4 b5"
             e(inst_1) : "l_inp1 l_inp2 l_inp3 l_lninp1 l_lninp2 l_lninp3 _cons"
         e(properties) : "b V"

matrices:
                  e(b) :  1 x 5
                  e(V) :  5 x 5
               e(init) :  1 x 5
       e(V_modelbased) :  5 x 5
                  e(S) :  7 x 7
                  e(W) :  7 x 7

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/