Plotting results using Coefplot

[Sherine Maui]

Hi Statalist,
I am trying to visually present my regression results. I would like to plot main effects as bar charts with error bars 95% confidence interval. This is the code I’ve run, it combines the results of three models in one graph:

xtreg M1  dbirthstat2 dsondum21 spercent2 totalchild i.educ i.releduc i.qwealth i.employment durban1 dlivingstructure1 age firstbirthage agegap agemarriage gdp1 i.round, re theta
est store Financial
xtreg D1  dbirthstat2 dsondum21 spercent2 totalchild i.educ i.releduc i.qwealth i.employment durban1 dlivingstructure1 age firstbirthage agegap agemarriage gdp1 i.round, re theta
est store Daily
xtreg C1  dbirthstat2 dsondum21 spercent2 totalchild i.educ i.releduc i.qwealth i.employment durban1 dlivingstructure1 age firstbirthage agegap agemarriage gdp1 i.round, re theta
est store Children
coefplot Financial Daily Children, vertical recast(bar) barwidth(0.25) fcolor(*.5) ciopts(recast(rcap)) citop  ytitle(Linear prediction) keep (dbirthstat2 dsondum21)

This code works fine (included as attachment). However, when I run the code below to plot the coefficents for educ and releduc I get the following error message

coefplot Financial Daily Children, vertical recast(bar) barwidth(0.25) fcolor(*.5) ciopts(recast(rcap)) citop  ytitle(Linear prediction) keep (educ releduc)
(Financial: no coefficients found, all dropped, or none kept)
(Daily: no coefficients found, all dropped, or none kept)
(Children: no coefficients found, all dropped, or none kept)
(nothing to plot)

I have not dropped my model estimates so I am not sure why there is nothing to plot

[Scott Merryman]

Because neither educ nor releduc are in your estimation commands. You have i.year and i.releduc

If you use -xtreg, coefleg- it will show the names of the coefficients. You can reference them in -coefplot with keep(*year *releduc). For example

. webuse grunfeld,clear

. qui xtreg invest mvalue kstock i.year, fe

. est store A

. xtreg , coefleg

Fixed-effects (within) regression               Number of obs     =        200
Group variable: company                         Number of groups  =         10

R-sq:                                           Obs per group:
     within  = 0.7985                                         min =         20
     between = 0.8143                                         avg =       20.0
     overall = 0.8068                                         max =         20

                                                F(21,169)         =      31.90
corr(u_i, Xb)  = -0.3250                        Prob > F          =     0.0000

------------------------------------------------------------------------------
      invest |      Coef.  Legend
-------------+----------------------------------------------------------------
      mvalue |   .1177158  _b[mvalue]
      kstock |   .3579163  _b[kstock]
             |
        year |
       1936  |  -19.19741  _b[1936.year]
       1937  |  -40.69001  _b[1937.year]
       1938  |   -39.2264  _b[1938.year]
       1939  |  -69.47029  _b[1939.year]
       1940  |  -44.23507  _b[1940.year]
       1941  |  -18.80446  _b[1941.year]
       1942  |  -21.13979  _b[1942.year]
       1943  |  -42.97762  _b[1943.year]
       1944  |  -43.09876  _b[1944.year]
       1945  |  -55.68303  _b[1945.year]
       1946  |  -31.16928  _b[1946.year]
       1947  |  -39.39223  _b[1947.year]
       1948  |  -43.71651  _b[1948.year]
       1949  |   -73.4951  _b[1949.year]
       1950  |  -75.89611  _b[1950.year]
       1951  |   -62.4809  _b[1951.year]
       1952  |  -64.63233  _b[1952.year]
       1953  |  -67.71796  _b[1953.year]
       1954  |  -93.52622  _b[1954.year]
             |
       _cons |  -32.83631  _b[_cons]
-------------+----------------------------------------------------------------
     sigma_u |  91.798268
     sigma_e |  51.724523
         rho |  .75902159   (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(9, 169) = 52.36                     Prob > F = 0.0000

. coefplot A, vertical recast(bar) keep(*year) /// 
>         barwidth(0.25) fcolor(*.5) ciopts(recast(rcap)) citop

[Sherine Maui]

Thank you Scott, the clears it up for me.