Modifying coefficient in Coefplot

[Atif Anwar]

Hello, I am running this regression and want to plot the coefficients with significance star and confidence interval.

xtreg HCR_URP c.ln_pc_nsdp_2##ib1.State Year, fe vce(robust)

For interpretation, I must add the log_pc_nsdp2 coefficient(Base state) with each state's coefficient.
Is it possible to plot the coefficients after adding.

Thank You

[Dimitriy V. Masterov]

You can use margins to do this:

. webuse pig, clear
(Longitudinal analysis of pig weights)

. xtset id week

Panel variable: id (strongly balanced)
 Time variable: week, 1 to 9
         Delta: 1 unit

. gen group = mod(id,3)

. reg weight c.week##i.group, vce(robust) // coeflegend

Linear regression                               Number of obs     =        432
                                                F(5, 426)         =    1317.15
                                                Prob > F          =     0.0000
                                                R-squared         =     0.9342
                                                Root MSE          =     4.2948

------------------------------------------------------------------------------
             |               Robust
      weight | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
        week |   6.040104   .1542765    39.15   0.000     5.736866    6.343342
             |
       group |
          1  |  -.4053819   .8390841    -0.48   0.629    -2.054642    1.243878
          2  |   .2821181   .9709362     0.29   0.772    -1.626304     2.19054
             |
group#c.week |
          1  |   .1442708   .1861637     0.77   0.439     -.221643    .5101846
          2  |   .3651042   .2269421     1.61   0.108    -.0809615    .8111698
             |
       _cons |    19.3967   .6640944    29.21   0.000     18.09139    20.70201
------------------------------------------------------------------------------

. margins group, dydx(week)

Average marginal effects                                   Number of obs = 432
Model VCE: Robust

Expression: Linear prediction, predict()
dy/dx wrt:  week

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
week         |
       group |
          0  |   6.040104   .1542765    39.15   0.000     5.736866    6.343342
          1  |   6.184375   .1041907    59.36   0.000     5.979583    6.389167
          2  |   6.405208   .1664376    38.48   0.000     6.078067     6.73235
------------------------------------------------------------------------------

. /* Verify By Hand */
. lincom _b[week] + 0 

 ( 1)  week = 0

------------------------------------------------------------------------------
      weight | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         (1) |   6.040104   .1542765    39.15   0.000     5.736866    6.343342
------------------------------------------------------------------------------

. lincom _b[week]+_b[1.group#c.week]

 ( 1)  week + 1.group#c.week = 0

------------------------------------------------------------------------------
      weight | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         (1) |   6.184375   .1041907    59.36   0.000     5.979583    6.389167
------------------------------------------------------------------------------

. lincom _b[week]+_b[2.group#c.week]

 ( 1)  week + 2.group#c.week = 0

------------------------------------------------------------------------------
      weight | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         (1) |   6.405208   .1664376    38.48   0.000     6.078067     6.73235
------------------------------------------------------------------------------

If you want to replicate lincom in your example, you can uncomment the coeflegend option to see how Stata names the coefficients.

[Atif Anwar]

Thanks for sharing but this is not helping,