Apply multiple transformations to different models (estout)

[Krista Lane]

When making tables using esttab, I know it's possible to apply a transformation to just some models by adding an option like:

local scalefac = 100
esttab model1 model2 model3,
transform(@*`scalefac' `scalefac', pattern(1 0 0))

Is it possible to apply a different transformation to different models? For example, I'd like to do:

local scalefac = 100
local scalefac2 = 200
esttab model1 model2 model3,
transform(@*`scalefac' `scalefac', pattern(1 0 0))
transform(@*`scalefac2' `scalefac2', pattern(0 1 0))

Thank you!

[Andrew Musau]

estout is from SSC (FAQ Advice #12). I think -transform()- allows exactly one transformation, but I may be wrong. You could manipulate the matrix and repost using erepost from SSC.

sysuse auto, clear
reg mpg weight  
mat b= e(b)*100
mat V= e(V)*100^2
erepost b=b, rename
erepost V=V
est sto m1

eststo m2: reg mpg weight disp 
eststo m3: reg mpg weight disp turn
local scalefac2 = 200
esttab m*, transform(@*`scalefac2' `scalefac2', pattern(0 1 0))

Res.:

. esttab m*, transform(@*`scalefac2' `scalefac2', pattern(0 1 0))

------------------------------------------------------------
                      (1)             (2)             (3)   
                      mpg             mpg             mpg   
------------------------------------------------------------
weight             -0.601***       -1.313***     -0.00593***
                 (-11.60)         (-5.63)         (-4.14)   

displacement                        1.056         0.00560   
                                   (0.54)          (0.56)   

turn                                               -0.139   
                                                  (-0.77)   

_cons              3944.0***       8016.9***        43.59***
                  (24.44)         (19.84)          (8.78)   
------------------------------------------------------------
N                      74              74              74   
------------------------------------------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001

[Krista Lane]

This worked perfectly. Thank you!