storing multiple beta coefficients for coefplot

[lion jj]

hi,

I would like to make a coefplot with betá's out of my regress split by a Dummy. like this:



i came across this post: https://www.statalist.org/forums/for...s-for-coefplot

This is only working for a singel regress. how to store storing beta's by groups?


sysuse auto, clear regress price mpg weight i.rep78, beta qui esttab, beta mat beta= e(beta) coefplot mat(beta), drop(_cons) scheme(s1mono)


can anyone help?

[lion jj]

please anyone?

[Andrew Musau]

hi,

I would like to make a coefplot with betá's out of my regress split by a Dummy. like this:

Like how? I do not see an example. coefplot is from the Stata Journal (FAQ Advice #12). Also, read and act on https://www.statalist.org/forums/help#realnames

[lion jj]

I'm sorry, i'm new too this form.

a exempel plot should be someting like;

Code: 
 regress studentsat totalincome teachingincome researchincome teachingstudentfte i.department i.region, beta   estimates store m1    coefplot m1, ///     plotlabels("Model 1" ) ///     title("Regression Coeffients Modelling Points")  ///     graphregion(fcolor(white))

[Andrew Musau]

If you need the confidence intervals in addition to the coefficients, it is easier to standardize the variables yourself prior to running the regressions. Standardizing involves subtracting the mean from the variable and dividing by its standard deviation.

sysuse auto
regress mpg weight disp, beta
preserve
foreach var in mpg weight  disp{
    qui sum `var'
    replace `var'= (`var'- r(mean))/r(sd)
}
regress mpg weight disp, nocons

Res.:

. regress mpg weight disp, beta

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(2, 71)        =     66.79
       Model |  1595.40969         2  797.704846   Prob > F        =    0.0000
    Residual |  848.049768        71  11.9443629   R-squared       =    0.6529
-------------+----------------------------------   Adj R-squared   =    0.6432
       Total |  2443.45946        73  33.4720474   Root MSE        =    3.4561

------------------------------------------------------------------------------
         mpg |      Coef.   Std. Err.      t    P>|t|                     Beta
-------------+----------------------------------------------------------------
      weight |  -.0065671   .0011662    -5.63   0.000                -.8821898
displacement |   .0052808   .0098696     0.54   0.594                 .0838253
       _cons |   40.08452    2.02011    19.84   0.000                        .
------------------------------------------------------------------------------


. 
. regress mpg weight disp, nocons

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(2, 72)        =     67.73
       Model |  47.6639405         2  23.8319703   Prob > F        =    0.0000
    Residual |   25.336059        72  .351889709   R-squared       =    0.6529
-------------+----------------------------------   Adj R-squared   =    0.6433
       Total |  72.9999996        74  .986486481   Root MSE        =     .5932

------------------------------------------------------------------------------
         mpg |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      weight |  -.8821898   .1555747    -5.67   0.000    -1.192322   -.5720572
displacement |   .0838253   .1555747     0.54   0.592    -.2263073    .3939579
------------------------------------------------------------------------------

For different models and coefplot, the code will be something like

sysuse auto, clear
preserve
foreach var in price mpg weight turn disp{
    qui sum `var'
    replace `var'= (`var'- r(mean))/r(sd)
}
eststo all: regress price mpg weight turn disp, nocons
restore, preserve
foreach var in price mpg weight turn disp{
    qui sum `var' if foreign
    replace `var'= (`var'- r(mean))/r(sd) if foreign
}
eststo foreign: regress price mpg weight turn disp if foreign, nocons
restore, preserve
foreach var in price mpg weight turn disp{
    qui sum `var' if !foreign
    replace `var'= (`var'- r(mean))/r(sd) if !foreign
}
eststo domestic: regress price mpg weight turn disp if !foreign, nocons
restore
coefplot all domestic foreign, scheme(s1color) leg(row(1))

Res.: