How to get the regression results as per country or region by using xtreg with re ?

[Sumran Ali]

I am using the below-mentioned command to get the results country wise, but I am not getting any results. Kindly help me with what I can do to get the regression results per country or region?


"by Crossid: xtreg GFSI lnOA lnFI gdp lnEmi ENG FERT lnERF lnANW Con GPCPI, re "

[Andrew Musau]

If you data is at the country level, you cannot do panel regression for each country. But you may for regions, provided that each region has more than one country. Here is an example using the Grunfeld dataset:

webuse grunfeld, clear
gen top5= company<=5
bys top5: xtreg invest mvalue kstock

Res.:

. bys top5: xtreg invest mvalue kstock

---------------------------------------------------------------------------------------------------------------------------------------
-> top5 = 0

Random-effects GLS regression                   Number of obs     =        100
Group variable: company                         Number of groups  =          5

R-sq:                                           Obs per group:
     within  = 0.7874                                         min =         20
     between = 0.6835                                         avg =       20.0
     overall = 0.7213                                         max =         20

                                                Wald chi2(2)      =     352.97
corr(u_i, X)   = 0 (assumed)                    Prob > chi2       =     0.0000

------------------------------------------------------------------------------
      invest |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      mvalue |   .0815434   .0075629    10.78   0.000     .0667203    .0963665
      kstock |   .1147979   .0133422     8.60   0.000     .0886477    .1409481
       _cons |  -7.223386   7.490499    -0.96   0.335    -21.90449    7.457722
-------------+----------------------------------------------------------------
     sigma_u |   15.81089
     sigma_e |  9.6545682
         rho |  .72840308   (fraction of variance due to u_i)
------------------------------------------------------------------------------

---------------------------------------------------------------------------------------------------------------------------------------
-> top5 = 1

Random-effects GLS regression                   Number of obs     =        100
Group variable: company                         Number of groups  =          5

R-sq:                                           Obs per group:
     within  = 0.7817                                         min =         20
     between = 0.7466                                         avg =       20.0
     overall = 0.7567                                         max =         20

                                                Wald chi2(2)      =     345.53
corr(u_i, X)   = 0 (assumed)                    Prob > chi2       =     0.0000

------------------------------------------------------------------------------
      invest |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      mvalue |   .1148413   .0158498     7.25   0.000     .0837763    .1459062
      kstock |   .3203831   .0239996    13.35   0.000     .2733447    .3674215
       _cons |  -81.95815   73.45286    -1.12   0.265    -225.9231    62.00682
-------------+----------------------------------------------------------------
     sigma_u |  152.89289
     sigma_e |  71.914089
         rho |  .81884338   (fraction of variance due to u_i)
------------------------------------------------------------------------------

.

[Sumran Ali]

thank you so much Andrew Musau. I got it