xtreg y x1 x2....,fe
estimates store fe
xtreg y x1 x2...., re

hausman fe


下面是我的一个例子 不过数据没法传上来 只能证明我的命令能用。

HAUSMAN TEST

. xtreg  ltrade lrgdp lrgdppc ldist comlang border regional landl island lareap comcol curcol colony comctry

>  custrict,fe

 

Fixed-effects (within) regression               Number of obs      =    219558

Group variable (i): pairid                      Number of groups   =     11178

 

R-sq:  within  = 0.1174                         Obs per group: min =         1

       between = 0.2289                                        avg =      19.6

       overall = 0.2236                                        max =        50

 

                                                F(5,208375)        =   5544.92

corr(u_i, Xb)  = 0.1438                         Prob > F           =    0.0000

 

------------------------------------------------------------------------------

      ltrade |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]

-------------+----------------------------------------------------------------

       lrgdp |   .0479512   .0087746     5.46   0.000     .0307531    .0651493

     lrgdppc |   .7909501   .0140174    56.43   0.000     .7634762    .8184239

       ldist |  (dropped)

     comlang |  (dropped)

      border |  (dropped)

    regional |    .692351   .0450073    15.38   0.000     .6041379    .7805642

       landl |  (dropped)

      island |  (dropped)

      lareap |  (dropped)

      comcol |  (dropped)

      curcol |    .358431   .0881309     4.07   0.000     .1856965    .5311654

      colony |  (dropped)

     comctry |  (dropped)

    custrict |   .6512318    .050578    12.88   0.000     .5521001    .7503634

       _cons |  -4.956948   .2282823   -21.71   0.000    -5.404376    -4.50952

-------------+----------------------------------------------------------------

     sigma_u |  3.3401212

     sigma_e |  1.3220505

         rho |  .86455466   (fraction of variance due to u_i)

------------------------------------------------------------------------------

F test that all u_i=0:     F(11177, 208375) =    34.50       Prob > F = 0.0000

 

. estimates store fe

 

. xtreg  ltrade lrgdp lrgdppc ldist comlang border regional landl island lareap comcol curcol colony comctry

>  custrict, re

 

Random-effects GLS regression                   Number of obs      =    219558

Group variable (i): pairid                      Number of groups   =     11178

 

R-sq:  within  = 0.1152                         Obs per group: min =         1

       between = 0.5207                                        avg =      19.6

       overall = 0.4656                                        max =        50

 

Random effects u_i ~ Gaussian                   Wald chi2(14)      =  40688.72

corr(u_i, X)       = 0 (assumed)                Prob > chi2        =    0.0000

 

------------------------------------------------------------------------------

      ltrade |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

-------------+----------------------------------------------------------------

       lrgdp |   .2694305   .0072176    37.33   0.000     .2552843    .2835767

     lrgdppc |   .5252985    .011516    45.61   0.000     .5027275    .5478695

       ldist |  -1.345633   .0314899   -42.73   0.000    -1.407352   -1.283914

     comlang |   .1760738   .0619909     2.84   0.005     .0545739    .2975736

      border |   .5309711   .1644242     3.23   0.001     .2087057    .8532366

    regional |   .6610592   .0448452    14.74   0.000     .5731643    .7489541

       landl |  -.8566298   .0425414   -20.14   0.000    -.9400094   -.7732501

      island |  -.0574684   .0517478    -1.11   0.267    -.1588923    .0439555

      lareap |   .2560334   .0091301    28.04   0.000     .2381387    .2739281

      comcol |  -.2770005   .0791023    -3.50   0.000    -.4320381   -.1219628

      curcol |   .4283967   .0887594     4.83   0.000     .2544316    .6023619

      colony |   3.208207   .2055153    15.61   0.000     2.805404     3.61101

     comctry |   1.272358   1.593654     0.80   0.425    -1.851146    4.395863

    custrict |    .700997   .0500125    14.02   0.000     .6029744    .7990196

       _cons |  -7.236036   .3224152   -22.44   0.000    -7.867958   -6.604114

-------------+----------------------------------------------------------------

     sigma_u |  2.1970043

     sigma_e |  1.3220536

         rho |  .73415697   (fraction of variance due to u_i)

------------------------------------------------------------------------------

 

. hausman fe

 

                 ---- Coefficients ----

             |      (b)          (B)            (b-B)     sqrt(diag(V_b-V_B))

             |       fe           .          Difference          S.E.

-------------+----------------------------------------------------------------

       lrgdp |    .0479512     .2694305       -.2214793        .0049901

     lrgdppc |    .7909501     .5252985        .2656516        .0079919

    regional |     .692351     .6610592        .0312918        .0038165

      curcol |     .358431     .4283967       -.0699658               .

    custrict |    .6512318      .700997       -.0497653        .0075425

------------------------------------------------------------------------------

                           b = consistent under Ho and Ha; obtained from xtreg

            B = inconsistent under Ha, efficient under Ho; obtained from xtreg

 

    Test:  Ho:  difference in coefficients not systematic

 

                  chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)

                          =     6418.73

                Prob>chi2 =      0.0000

                (V_b-V_B is not positive definite)

在本版搜索“hausman”

2楼的说明很详细

好人，刚刚学习的菜鸟受用了

谢谢啊。。。。。