ESTTAB Command Help

[Davia Downey]

I have 4 models (OLS, Fixed Effects, Random Effects and a Fixed Effect Differenced) which I have stored using the eststo command and am trying to use esttab to get certain scalars to be reported as an output table so that I can format the resulting table for an article. I am interested in the following scalars r2 df_2 F Sigma u,the Rho statistic and the standard errors of the estimates (SEE) or square root of mean squared errors (SRMSE).

The current code I'm using (which only gets me part of the way) is

esttab, label mtitles scalars (F p_value  df_r  r2)

I am unsure how to find the scalar equivalent to SEE, SRMSE, Rho, the actual F-test(model) and how to add stars to the F-test and Effect Test in the output. Please advise. Should I simply run a display after each model and record it so that I can edit the table manually or is there any easy way to get this using the scalar function?

Fixed/Random Effects Statistics Highlighted in Bold:

. xtreg $ylist l.lnannualjobs d.pct_aanh pct_poverty pct_totaldegrees d.ln_empstable avgstateincentive, fe i(county)

Fixed-effects (within) regression               Number of obs     =        555
Group variable: county                          Number of groups  =         37

R-sq:                                           Obs per group:
     within  = 0.4762                                         min =         15
     between = 0.9936                                         avg =       15.0
     overall = 0.9787                                         max =         15

                                                F(6,512)          =      77.58
corr(u_i, Xb)  = 0.9691                         Prob > F          =     0.0000

-----------------------------------------------------------------------------------
     lnannualjobs |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
     lnannualjobs |
              L1. |   .5567163   .0347663    16.01   0.000     .4884142    .6250184
                  |
         pct_aanh |
              D1. |   .0010539   .0020719     0.51   0.611    -.0030166    .0051243
                  |
      pct_poverty |   .0129627    .003774     3.43   0.001     .0055482    .0203772
 pct_totaldegrees |   .0129081   .0067972     1.90   0.058    -.0004457    .0262619
                  |
     ln_empstable |
              D1. |   .7517452   .0451105    16.66   0.000     .6631207    .8403697
                  |
avgstateincentive |  -2.518008   5.451199    -0.46   0.644    -13.22748    8.191461
            _cons |   3.582079   .3671468     9.76   0.000     2.860779    4.303379
------------------+----------------------------------------------------------------
          sigma_u |  .54256832
          sigma_e |  .13076773
              rho |  .94510027   (fraction of variance due to u_i)
-----------------------------------------------------------------------------------
F test that all u_i=0: F(36, 512) = 5.14                     Prob > F = 0.0000

OLS Statistics Needed Highlighted in Bold

reg $ylist $xlist if obs!=219

      Source |       SS           df       MS      Number of obs   =       591
-------------+----------------------------------   F(5, 585)       =    213.57
       Model |  592.562504         5  118.512501   Prob > F        =    0.0000
    Residual |  324.625907       585  .554916081   R-squared       =    0.6461
-------------+----------------------------------   Adj R-squared   =    0.6430
       Total |  917.188411       590  1.55455663   Root MSE        =    .74493

-----------------------------------------------------------------------------------
     lnannualjobs |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
         pct_aanh |  -.0086078   .0026004    -3.31   0.001     -.013715   -.0035006
      pct_poverty |     .01679   .0085872     1.96   0.051    -.0000755    .0336555
 pct_totaldegrees |  -.0146091   .0230061    -0.64   0.526    -.0597936    .0305755
     ln_empstable |   .8816694   .0474711    18.57   0.000     .7884349    .9749039
avgstateincentive |  -30.33077   25.47656    -1.19   0.234    -80.36743    19.70589
            _cons |   .9807619   .9065059     1.08   0.280    -.7996406    2.761164
-----------------------------------------------------------------------------------

Apologies for the long post, I wanted to provide as much detail as possible. Many thanks in advance.

[Andrew Musau]

esttab is from Stata Journal. Most of these statistics will be available from ereturn list. If not, you can reproduce them, so there will be no need to insert anything manually.

sysuse auto
qui regress mpg weight disp
ereturn list

Res.:

scalars:
                  e(N) =  74
               e(df_m) =  2
               e(df_r) =  71
                  e(F) =  66.78504752026517
                 e(r2) =  .6529306984682528
               e(rmse) =  3.45606176570828
                e(mss) =  1595.409691543724
                e(rss) =  848.0497679157351
               e(r2_a) =  .643154098425105
                 e(ll) =  -195.2397979466294
               e(ll_0) =  -234.3943376482347
               e(rank) =  3

So in the esttab command, specify

esttab, stats(F rmse rss, star(F))

Res.:

. esttab, stats(F rmse rss, star(F))

----------------------------
                      (1)   
                      mpg   
----------------------------
weight           -0.00657***
                  (-5.63)   

displacement      0.00528   
                   (0.54)   

_cons               40.08***
                  (19.84)   
----------------------------
F                   66.79***
rmse                3.456   
rss                 848.0   
----------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001

Similarly, you can get your wanted statistics from xtreg.

[Davia Downey]

Thanks, Andrew. I knew it was something like that. I read everything about making tables using esttab, but didn’t consider looking at the stored scalars in xtreg for my answer.