order of independent variables in -esttab- command

[Max Piper]

the following code shows two "interesting" independent variables (mpg and displacement) whose impact on price I want to investigate with two control variables length and weight.

sysuse auto.dta, clear

estimates clear
reg price mpg weight length
estimates store reg1
reg price displacement weight length
estimates store reg2

esttab *

The output of this esttab is, however:

--------------------------------------------
                      (1)             (2)  
                    price           price  
--------------------------------------------
mpg                -86.79                  
                  (-1.03)                  

weight              4.365***        4.613**
                   (3.74)          (3.30)  

length             -104.9*         -97.63*  
                  (-2.64)         (-2.47)  

displacement                        0.727  
                                   (0.10)  

_cons             14542.4*        10440.6*  
                   (2.47)          (2.39)  
--------------------------------------------
N                      74              74  
--------------------------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001

As you can see, the variables mpg and displacement are visually separated, counterintuitive to the content and interpretation (this is a valid concern, no?). What can I do to have these models

• in the same table
• with the variables that I identify as "interesting", but that are not in each of the models, at the top

i.e. like this:

--------------------------------------------
                      (1)             (2)  
                    price           price  
--------------------------------------------
mpg                -86.79                  
                  (-1.03)                  

displacement                        0.727  
                                   (0.10)  

weight              4.365***        4.613**
                   (3.74)          (3.30)  

length             -104.9*         -97.63*  
                  (-2.64)         (-2.47)  

_cons             14542.4*        10440.6*  
                   (2.47)          (2.39)  
--------------------------------------------
N                      74              74  
--------------------------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001

I am aware that I can put them next to each other at the bottom of the table, by typing

sysuse auto.dta, clear

estimates clear
reg price weight length mpg
estimates store reg1
reg price weight length displacement
estimates store reg2

esttab *

which gives

--------------------------------------------
                      (1)             (2)  
                    price           price  
--------------------------------------------
weight              4.365***        4.613**
                   (3.74)          (3.30)  

length             -104.9*         -97.63*  
                  (-2.64)         (-2.47)  

mpg                -86.79                  
                  (-1.03)                  

displacement                        0.727  
                                   (0.10)  

_cons             14542.4*        10440.6*  
                   (2.47)          (2.39)  
--------------------------------------------
N                      74              74  
--------------------------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001

However, I would find it very odd to find these interesting variables wedged between the control variables and the number of observations, r^2, etc. (no?), so what can I do?

[William Lisowski]

The output of help esttab points us to the order option to achieve precisely specified ordering of the results.

. esttab *, order(mpg displacement)

--------------------------------------------
                      (1)             (2)  
                    price           price  
--------------------------------------------
mpg                -86.79                  
                  (-1.03)                  

displacement                        0.727  
                                   (0.10)  

weight              4.365***        4.613**
                   (3.74)          (3.30)  

length             -104.9*         -97.63*  
                  (-2.64)         (-2.47)  

_cons             14542.4*        10440.6*  
                   (2.47)          (2.39)  
--------------------------------------------
N                      74              74  
--------------------------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001