Formatting test of means table

[Kayleigh Amber]

Hello all,

I would like to recreate as much as possible the following table (wih different variables):


I have managed to get most of the way using the following code:


. do "C:\Users\Kris\AppData\Local\Temp\STD5320_000000.t mp"

. eststo lessthan15: quietly estpost summarize SALES_REV_TURN-SHORT_AND_LONG_TERM_DEBT if ave15 == 0, detail

. eststo morethan15: quietly estpost summarize SALES_REV_TURN-SHORT_AND_LONG_TERM_DEBT if ave15 == 1, detail

. eststo diff: quietly estpost ttest SALES_REV_TURN-SHORT_AND_LONG_TERM_DEBT, by(ave15) unequal

. esttab lessthan15 morethan15 diff, cells("mean(pattern(1 1 0) fmt(3)) sd(pattern(1 1 0)) median(pattern(1 1 0)) b(star pattern(0 0 1) fmt(3)) t(patt
> ern(0 0 1) par fmt(3))")


However, as you can see from my results I am missing column names for 1, 2 and 3, as well as a title line, and explanatory line as above. I am also missing the median.



Any help would be greatly appreciated,
Kayleigh

[Andrew Musau]

estout is from SSC (FAQ Advice #12). If you remove the -quietly- prefix, you can see how estout labels the statistics. The median is the 50th percentile and is stored as "p50".

sysuse auto, clear
eststo lessthan15: quietly estpost summarize mpg weight turn if !foreign, detail
eststo morethan15: quietly estpost summarize mpg weight turn if foreign, detail
eststo diff: quietly estpost ttest mpg weight turn, by(foreign) unequal

esttab lessthan15 morethan15 diff, label collab(, lhs("Variables")) ///
cells("mean(pattern(1 1 0) fmt(3) label(Mean)) sd(pattern(1 1 0) label(SD)) p50(pattern(1 1 0) label(Median) fmt(a3)) b(star pattern(0 0 1) fmt(3) label(Diff.)) t(pattern(0 0 1) par fmt(3) label(t-stat.))")  ///
title("T-test of comparisons of means`=char(10)'Some additional information") nonumbers mtitles("Group 1" "Group 2" "Comparison")

Res.:

T-test of comparisons of means
Some additional information
-------------------------------------------------------------------------------------------------------------------------------
                          Group 1                                Group 2                             Comparison                
Variables                    Mean           SD       Median         Mean           SD       Median        Diff.         t-stat.
-------------------------------------------------------------------------------------------------------------------------------
Mileage (mpg)              19.827        4.743           19       24.773        6.611        24.50       -4.946**      (-3.180)
Weight (lbs.)            3317.115      695.364         3360     2315.909      433.003         2180     1001.206***      (7.500)
Turn Circle (ft.)          41.442        3.968           42       35.409        1.501           36        6.033***      (9.479)
-------------------------------------------------------------------------------------------------------------------------------
Observations                   52                                     22                                     74                
-------------------------------------------------------------------------------------------------------------------------------

[Kayleigh Amber]

Thanks again Andrew! This worked perfectly

[Maysam Ayoub]

Hello, I used the command for a similar test that I am doing and it works perfectly; however, the table does not show the p-value
Can you pls let me know where i can add the pval command

[Andrew Musau]

sysuse auto, clear
eststo lessthan15: quietly estpost summarize mpg weight turn if !foreign, detail
eststo morethan15: quietly estpost summarize mpg weight turn if foreign, detail
eststo diff: quietly estpost ttest mpg weight turn, by(foreign) unequal

esttab lessthan15 morethan15 diff, label collab(, lhs("Variables")) ///
cells("mean(pattern(1 1 0) fmt(3) label(Mean)) sd(pattern(1 1 0)label(SD)) p50(pattern(1 1 0) label(Median) fmt(a3)) b(pattern(0 0 1) fmt(3) label(Diff.)) t(pattern(0 0 1) par fmt(3) label(t-stat.)) p(pattern(0 0 1)fmt(3) label(p-value))")  ///
title("T-test of comparisons of means`=char(10)'Some additional information") ///
nonumbers mtitles("Group 1" "Group 2" "Comparison") substitute(0.000 "<0.001") 

Res.:

. esttab lessthan15 morethan15 diff, label collab(, lhs("Variables")) ///
> cells("mean(pattern(1 1 0) fmt(3) label(Mean)) sd(pattern(1 1 0)label(SD)) p50(pattern(1 1 0) label(Median) fmt(a3)) b(pattern(0 0 1) fmt(3) 
> label(Diff.)) t(pattern(0 0 1) par fmt(3) label(t-stat.)) p(pattern(0 0 1)fmt(3) label(p-value))")  ///
> title("T-test of comparisons of means`=char(10)'Some additional information") ///
> nonumbers mtitles("Group 1" "Group 2" "Comparison") substitute(0.000 "<0.001") 

T-test of comparisons of means
Some additional information
-----------------------------------------------------------------------------------------------------------------------------------------
                          Group 1                                Group 2                             Comparison                          
Variables                    Mean           SD       Median         Mean           SD       Median        Diff.      t-stat.      p-value
-----------------------------------------------------------------------------------------------------------------------------------------
Mileage (mpg)              19.827        4.743           19       24.773        6.611        24.50       -4.946     (-3.180)        0.003
Weight (lbs.)            3317.115      695.364         3360     2315.909      433.003         2180     1001.206      (7.500)        <0.001
Turn Circle (ft.)          41.442        3.968           42       35.409        1.501           36        6.033      (9.479)        <0.001
-----------------------------------------------------------------------------------------------------------------------------------------
Observations                   52                                     22                                     74                          
-----------------------------------------------------------------------------------------------------------------------------------------

[Maysam Ayoub]

thank you so much!