Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • How to test the significance of difference in coefficients of the same explanatory variable in two different 3sls regressions

    Dear Statalist users,

    Hope you all are doing pretty well. I am currently trying to test the significance of difference in coefficients of two different 3sls regressions, given same explanatory variables in these two regressions but different sub-samples. I have run these two regressions using 3sls in Stata, and used the following commands to test the difference of coefficients of these two regressions:

    reg3 (os age_hh edu_h perob sos ecs ens) (sos edu_hh mem nhi dtvac persob pare proe rece dise swfc swril) (ecs gen_hh age_hh mem hs dtvac pomm fir perecb pare proe rece dise swp swntfp swofi) (ens age_hh edu_hh dtrl nhi perenb pare proe rece dise swsem swib swwqi) if poverty == 1

    estimates store povertyone

    reg3 (os age_hh edu_h perob sos ecs ens) (sos edu_hh mem nhi dtvac persob pare proe rece dise swfc swril) (ecs gen_hh age_hh mem hs dtvac pomm fir perecb pare proe rece dise swp swntfp swofi) (ens age_hh edu_hh dtrl nhi perenb pare proe rece dise swsem swib swwqi) if poverty == 0

    estimates store povertyzero

    suest povertyone povertyzero

    test [povertyone_mean]age_hh = [povertyzero_mean]age)hh

    However, Stata 17 shows that reg3 is not supported by suest. I was thinking whether there is an alternative to the 'suest' command. Or is there any other way of testing the difference of coefficients of the above two different regressions? I would appreciate it of you could kindly give me a hint or advice on how to resolve the issue. Thanks a lot in advance!

    Have a nice day,
    Lingling



Working...
X