output svy results to excel

[Caroline Wilson]

HI all, I'd like to export the below results (specifically the column for RACETHM_n, proportion, lb and ub) to excel. It seems there are many options such as putexcel, tabout, esttab. Could someone suggest some code I could try to get these into excel?

Many thanks in advance for any advice.

.  svy, subpop(DGRDG_n1): tabulate RACETHM_n, ci level(90)
(running tabulate on estimation sample)

Number of strata   =         1                  Number of obs     =      1,311
                                                Population size   = 252,142.35
                                                Subpop. no. obs   =        547
                                                Subpop. size      = 113,793.95
                                                Replications      =        320
                                                Design df         =        319

----------------------------------------------
RACETHM_n | proportion          lb          ub
----------+-----------------------------------
 AsianNHO |       .078       .0527       .1138
     AIAN |    7.3e-05     1.0e-05     5.1e-04
    Black |      .0423        .018       .0965
 Hispanic |      .0695       .0368       .1274
    White |      .7969       .7237       .8546
       MR |      .0132       .0056       .0306
          | 
    Total |          1                        
----------------------------------------------
  Key:  proportion  =  cell proportion
        lb          =  lower 90% confidence bound for cell proportion
        ub          =  upper 90% confidence bound for cell proportion

[Andrew Musau]

estout from SSC works really well with -svy- commands.

ssc install estout, replace
eststo m1: estpost svy, subpop(DGRDG_n1): tabulate RACETHM_n, ci level(90)
esttab m1 using myfile.csv, replace cells("b lb ub")  collab(proportion lb ub) nonum mlab(none)

[Caroline Wilson]

This worked perfectly. Thank you so much!!

[Caroline Wilson]

Just a followup question. Is there a way to round the output for proportion and the CIs to 3 decimal places?

[Andrew Musau]

You can use:

esttab m1 using myfile.csv, replace cells("b(fmt(3)) lb(fmt(3)) ub(fmt(3))") collab(proportion lb ub) nonum mlab(none)

although I'd recommend

esttab m1 using myfile.csv, replace cells("b(fmt(a3)) lb(fmt(a3)) ub(fmt(a3))") collab(proportion lb ub) nonum mlab(none)

so that you avoid integers like 1 being written as "1.000".

[Caroline Wilson]

Excellent. thanks again!!!