Way to obtain std.error of interaction term

[Marco Lazzeretti]

I have this linear regression with mixed effects but I have a problem with interpretation of coefficents:

 
 xtmixed ild c.tempomesi##i.trattato || pid:, mle nolog variance   Mixed-effects ML regression                     Number of obs      =       143 Group variable: pid                             Number of groups   =        61                                                  Obs per group: min =         2                                                                avg =       2.3                                                                max =         6                                                   Wald chi2(3)       =     28.76 Log likelihood =  -550.1897                     Prob > chi2        =    0.0000  --------------------------------------------------------------------------------------                  ild |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval] ---------------------+----------------------------------------------------------------            tempomesi |   .3300396   .0657651     5.02   0.000     .2011423    .4589369           1.trattato |  -3.826831   4.163937    -0.92   0.358      -11.988    4.334336                      | trattato#c.tempomesi |                   1  |  -.2629648   .1021319    -2.57   0.010    -.4631397   -.0627899                      |                _cons |   20.78178   3.101511     6.70   0.000     14.70293    26.86063 --------------------------------------------------------------------------------------  ------------------------------------------------------------------------------   Random-effects Parameters  |   Estimate   Std. Err.     [95% Conf. Interval] -----------------------------+------------------------------------------------ pid: Identity                |                   var(_cons) |   226.6245    44.9081      153.6852     334.181 -----------------------------+------------------------------------------------                var(Residual) |   43.09052   6.757185       31.6885    58.59517 ------------------------------------------------------------------------------ LR test vs. linear regression: chibar2(01) =    96.58 Prob >= chibar2 = 0.0000

Trattato is a binary variable, thus in a untreated group tempomesi is 0.3300396 and in treated group is 0.3300396-0.2629648. But which is the std.err. of treated coefficient? Can i obtain this in other way? Because i've tried in this way but i did not obtain the same coefficient:

. xtmixed ild tempomesi ||pid: if trattato==1, nolog variance mle

Mixed-effects ML regression                     Number of obs      =        79
Group variable: pid                             Number of groups   =        34

                                                Obs per group: min =         2
                                                               avg =       2.3
                                                               max =         4


                                                Wald chi2(1)       =      1.24
Log likelihood = -289.09181                     Prob > chi2        =    0.2649

------------------------------------------------------------------------------
         ild |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
   tempomesi |   .0691339   .0620092     1.11   0.265    -.0524019    .1906696
       _cons |   16.94165   2.405656     7.04   0.000     12.22665    21.65665
------------------------------------------------------------------------------

------------------------------------------------------------------------------
  Random-effects Parameters  |   Estimate   Std. Err.     [95% Conf. Interval]
-----------------------------+------------------------------------------------
pid: Identity                |
                  var(_cons) |   174.2725   45.38596      104.6043    290.3408
-----------------------------+------------------------------------------------
               var(Residual) |   27.01511   5.705192      17.85852    40.86654
------------------------------------------------------------------------------
LR test vs. linear regression: chibar2(01) =    63.17 Prob >= chibar2 = 0.0000

[Marco Lazzeretti]

The first code was this:

. xtmixed ild c.tempomesi##i.trattato || pid:, mle nolog variance 

Mixed-effects ML regression                     Number of obs      =       143
Group variable: pid                             Number of groups   =        61

                                                Obs per group: min =         2
                                                               avg =       2.3
                                                               max =         6


                                                Wald chi2(3)       =     28.76
Log likelihood =  -550.1897                     Prob > chi2        =    0.0000

--------------------------------------------------------------------------------------
                 ild |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
           tempomesi |   .3300396   .0657651     5.02   0.000     .2011423    .4589369
          1.trattato |  -3.826831   4.163937    -0.92   0.358      -11.988    4.334336
                     |
trattato#c.tempomesi |
                  1  |  -.2629648   .1021319    -2.57   0.010    -.4631397   -.0627899
                     |
               _cons |   20.78178   3.101511     6.70   0.000     14.70293    26.86063
--------------------------------------------------------------------------------------

------------------------------------------------------------------------------
  Random-effects Parameters  |   Estimate   Std. Err.     [95% Conf. Interval]
-----------------------------+------------------------------------------------
pid: Identity                |
                  var(_cons) |   226.6245    44.9081      153.6852     334.181
-----------------------------+------------------------------------------------
               var(Residual) |   43.09052   6.757185       31.6885    58.59517
------------------------------------------------------------------------------
LR test vs. linear regression: chibar2(01) =    96.58 Prob >= chibar2 = 0.0000

[Maarten Buis]

The effect (with standard error confidence interval etc.) of grade for non-union members is the main effect of grade. The effect (with standard error, confidence interval, etc.) of grade for union members is returned by lincom.

webuse nlswork, clear
xtmixed ln_w i.union##c.grade || id:
lincom grade + 1.union#c.grade

[Marco Lazzeretti]

Thank you Maarten