Hello
I run OLS for my panel data ( with N=816, T=8) and check heteroskedasticity as below :
Code:
  . reg To vl0 ll0

      Source |       SS           df       MS      Number of obs   =       816
-------------+----------------------------------   F(2, 813)       =    291.16
       Model |  1408.90316         2   704.45158   Prob > F        =    0.0000
    Residual |  1967.01621       813  2.41945413   R-squared       =    0.4173
-------------+----------------------------------   Adj R-squared   =    0.4159
       Total |  3375.91937       815  4.14223236   Root MSE        =    1.5555

------------------------------------------------------------------------------
          To |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         vl0 |   3.862433    1.41577     2.73   0.007     1.083438    6.641429
         ll0 |   6.818981   .2842538    23.99   0.000     6.261023    7.376939
       _cons |  -6.080591   1.634762    -3.72   0.000    -9.289442   -2.871739
------------------------------------------------------------------------------
Code:
.
estat hettest

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity 
         Ho: Constant variance
         Variables: fitted values of To

         chi2(1)      =    21.03
         Prob > chi2  =   0.0000
Because there is heteroskedasticity I try to run GLS to fix this as follow as
Code:
 . predict e, residual

. gen e2=e^2

. qui reg e2 vl0 ll0

. predict v
(option xb assumed; fitted values)

. regress To vl0 ll0 [aweight=1/v]
(sum of wgt is   3.6132e+02)

      Source |       SS           df       MS      Number of obs   =       816
-------------+----------------------------------   F(2, 813)       =    290.21
       Model |  1312.56218         2   656.28109   Prob > F        =    0.0000
    Residual |  1838.49921       813  2.26137664   R-squared       =    0.4165
-------------+----------------------------------   Adj R-squared   =    0.4151
       Total |  3151.06139       815    3.866333   Root MSE        =    1.5038

------------------------------------------------------------------------------
          To |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         vl0 |   3.147649   1.357049     2.32   0.021     .4839154    5.811383
         ll0 |   6.319949    .264015    23.94   0.000     5.801718    6.838181
       _cons |  -4.720035   1.558333    -3.03   0.003    -7.778866   -1.661205
------------------------------------------------------------------------------
Is this correct? and I also wonder is there any quicker command in Stata to run GLS in my case or not?
Thank in advance