[Andrew Musau]

The issue is with your variable "exp". Its values are too large. Rescale and reestimate the models.

replace exp= exp/1e+9
lab variable exp "Expenditure in billions"

[Clara Joan Joachim]

Andrew Musau thank you very much prof!

[Badiah Eljahimi]

Hi all, is it correct if you substitute the order of the estimator as below, and how can I solve the problem of (V_b-V_B is not positive definite) thanks a lot!

[Badiah Eljahimi]

Andrew Musau Hi, can you help with these results please, is it correct if you substitute the order of the estimator as below, and how can I solve the problem of (V_b-V_B is not positive definite) thanks a lot!

hausman DFE pmg, sigmamor

                 ---- Coefficients ----
             |      (b)          (B)            (b-B)     sqrt(diag(V_b-V_B))
             |      DFE          pmg         Difference          S.E.
-------------+----------------------------------------------------------------
         fdx |
         L1. |   -38.78499    -6.619715       -32.16528        2.436517
     IHS_gee |
         L1. |   -14.41552    -6.993847       -7.421678               .
      FDXgee |
         L1. |    23.76378     2.059334        21.70444        2.627698
     IHS_inf |   -1.313541    -1.748122        .4345803        .4123638
     IHS_trd |
         L1. |    .2484932    -2.947831        3.196324        1.953859
------------------------------------------------------------------------------
                           b = consistent under Ho and Ha; obtained from xtpmg
            B = inconsistent under Ha, efficient under Ho; obtained from xtpmg

    Test:  Ho:  difference in coefficients not systematic

                  chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                          =   -42.02    chi2<0 ==> model fitted on these
                                        data fails to meet the asymptotic
                                        assumptions of the Hausman test;
                                        see suest for a generalized test

 hausman pmg DFE, sigmamor

                 ---- Coefficients ----
             |      (b)          (B)            (b-B)     sqrt(diag(V_b-V_B))
             |      pmg          DFE         Difference          S.E.
-------------+----------------------------------------------------------------
         fdx |
         L1. |   -6.619715    -38.78499        32.16528               .
     IHS_gee |
         L1. |   -6.993847    -14.41552        7.421678        6.421948
      FDXgee |
         L1. |    2.059334     23.76378       -21.70444               .
     IHS_inf |   -1.748122    -1.313541       -.4345803               .
     IHS_trd |
         L1. |   -2.947831     .2484932       -3.196324               .
------------------------------------------------------------------------------
                           b = consistent under Ho and Ha; obtained from xtpmg
            B = inconsistent under Ha, efficient under Ho; obtained from xtpmg

    Test:  Ho:  difference in coefficients not systematic

                  chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                          =       78.49
                Prob>chi2 =      0.0000
                (V_b-V_B is not positive definite)

[Carlo Lazzaro]

Badiah:
no, it is not.

The sequence (that should not be reversed) is:
1) the estimator that is consitent under both H0 and H1 but inefficient if H0 is not rejected;
2) the estimator that is consistent and efficient if the null is not rejected but inconsistent if H0 is rejected.
Therefore:

 hausman DFE pmg, sigmamore

In addition, see last example, -hausman- entry, Stata .pdf manual, where the negative -hausman- is considered as a strong evidence that the null cannot be rejected.

[Badiah Eljahimi]

Thank you very much, that's really helpful!