Comparing coefficients from two regressions

[Tosca La Grand]

Hello,

For a school paper I am replicating a study on partner preferences. We have to replicate one particular table, which presents the coefficients of degree of attractiveness and education with reply for men and women separately. It also shows the p-value for the F-test of equality across gender, and I can't seem to figure out how that's done.


I did some research on this website and figured the command -suest- would do the trick for me, but I get an error when I do this:

reg reply t_uni t_med t_high if female==1, r
estimates store female
reg reply t_uni t_med t_high if female==0, r
estimates store male

suest male female

male was estimated with a nonstandard vce (robust)
r(322);


Does anyone have any advice?

[Joseph Coveney]

I get an error when I do this:

reg reply t_uni t_med t_high if female==1, r[obust]
estimates store female
reg reply t_uni t_med t_high if female==0, r[obust]
estimates store male

suest male female

male was estimated with a nonstandard vce (robust)
r(322);


Does anyone have any advice?

Sure: try avoiding a nonstandard vce (robust).

[Carlo Lazzaro]

Tosca:
the Technical note at page 2664, -sureg- entry, Stata .pdf manual can hopefully shed light on this issue.

[Xiaogeng Xu]

Hello,

For a school paper I am replicating a study on partner preferences. We have to replicate one particular table, which presents the coefficients of degree of attractiveness and education with reply for men and women separately. It also shows the p-value for the F-test of equality across gender, and I can't seem to figure out how that's done.

[ATTACH=CONFIG]temp_11272_1529924697472_340[/ATTACH]
I did some research on this website and figured the command -suest- would do the trick for me, but I get an error when I do this:

reg reply t_uni t_med t_high if female==1, r
estimates store female
reg reply t_uni t_med t_high if female==0, r
estimates store male

suest male female

male was estimated with a nonstandard vce (robust)
r(322);


Does anyone have any advice?

Try:
reg reply t_uni t_med t_high if female==1
estimates store female
reg reply t_uni t_med t_high if female==0
estimates store male

suest male female, vce(robust)