Dear community,
I have tried to find a solution to add significance stars to lincom combinations in esttab.
E.g:
sysuse auto
reg price c.mpg##foreign
lincom 1.foreign + 1.foreign#c.mpg
estadd scalar lincom = `r(estimate)'
estadd scalar lincom_se = `r(se)'
esttab, se replace b(2) ///
starlevel(* 0.10 ** 0.05 *** 0.01) ///
stat(lincom lincom_se, labels("1.foreign + 1.foreign#c.mpg" "S.E.") ///
layout(@ (@)) fmt(2 2) star(lincom) )
produces the following table:
----------------------------
(1)
price
----------------------------
mpg -329.26***
(74.99)
0.foreign 0.00
(.)
1.foreign -13.59
(2634.66)
0.foreign#~g 0.00
(.)
1.foreign#~g 78.89
(112.48)
_cons 12600.54***
(1527.89)
----------------------------
1.foreign ~g 65.30***
S.E. (2526.43)
----------------------------
Standard errors in parentheses
* p<0.10, ** p<0.05, *** p<0.01
but clearly something went wrong here, since the interaction effect is not supposed to be significant (i.e., using the star(lincom) option is obviously not the correct thing to do). Any suggestions on how to display correct significance stars for individuals coefficients?
Best,
Ingar
I have tried to find a solution to add significance stars to lincom combinations in esttab.
E.g:
sysuse auto
reg price c.mpg##foreign
lincom 1.foreign + 1.foreign#c.mpg
estadd scalar lincom = `r(estimate)'
estadd scalar lincom_se = `r(se)'
esttab, se replace b(2) ///
starlevel(* 0.10 ** 0.05 *** 0.01) ///
stat(lincom lincom_se, labels("1.foreign + 1.foreign#c.mpg" "S.E.") ///
layout(@ (@)) fmt(2 2) star(lincom) )
produces the following table:
----------------------------
(1)
price
----------------------------
mpg -329.26***
(74.99)
0.foreign 0.00
(.)
1.foreign -13.59
(2634.66)
0.foreign#~g 0.00
(.)
1.foreign#~g 78.89
(112.48)
_cons 12600.54***
(1527.89)
----------------------------
1.foreign ~g 65.30***
S.E. (2526.43)
----------------------------
Standard errors in parentheses
* p<0.10, ** p<0.05, *** p<0.01
but clearly something went wrong here, since the interaction effect is not supposed to be significant (i.e., using the star(lincom) option is obviously not the correct thing to do). Any suggestions on how to display correct significance stars for individuals coefficients?
Best,
Ingar
Comment