I would like to see if the regression coefficients are statistically different for a low income group vs a high income group, when looking at the reaction to different levels of cost sharing (insurance plans)
the ran_plan_groups are dummies for which insurance plan the participants belong to. pct50_income_person is a dummy as well assigning participants to the two income groups.
Is this the rigtig way to approach it? sureg was mentioned somewhere as well.
Code:
set seed 620549 local controls age baseline_fam_size good_health bad_health quietly regress log_spending rand_plan_group2 rand_plan_group3 rand_plan_group4 rand_plan_group5 rand_plan_group6 `controls' demeaned_fam_start_month_site* demeaned_cal_year* if pct50_income_person ==0 est sto model_low_income quietly regress log_spending rand_plan_group2 rand_plan_group3 rand_plan_group4 rand_plan_group5 rand_plan_group6 `controls' demeaned_fam_start_month_site* demeaned_cal_year* if pct50_income_person ==1 est sto model_high_income suest model_low_income model_high_income, cluster(ifamily)
I get the following results but im not sure how to interpret it.
Code:
Number of obs = 17,810 (Std. Err. adjusted for 2,973 clusters in ifamily) -------------------------------------------------------------------------------------- | Robust | Coef. Std. Err. z P>|z| [95% Conf. Interval] ---------------------+---------------------------------------------------------------- model_low_income_m~n | rand_plan_group2 | -.8133025 .1192149 -6.82 0.000 -1.046959 -.5796456 rand_plan_group3 | -.6605734 .1177767 -5.61 0.000 -.8914115 -.4297352 rand_plan_group4 | -1.030376 .1515841 -6.80 0.000 -1.327476 -.7332768 rand_plan_group5 | -1.441263 .1193879 -12.07 0.000 -1.675259 -1.207267 rand_plan_group6 | -1.087224 .0926923 -11.73 0.000 -1.268898 -.9055505 ---------------------+---------------------------------------------------------------- model_low_income_l~r | _cons | 1.588036 .0199406 79.64 0.000 1.548953 1.627118 ---------------------+----------------------------------------------------------------
Code:
model_high_income_~n | rand_plan_group2 | -.400343 .1259898 -3.18 0.001 -.6472784 -.1534076 rand_plan_group3 | -.2398797 .1964297 -1.22 0.222 -.6248748 .1451154 rand_plan_group4 | -.7494231 .1771273 -4.23 0.000 -1.096586 -.40226 rand_plan_group5 | -1.089795 .1532231 -7.11 0.000 -1.390107 -.7894836 rand_plan_group6 | -.6580158 .1334854 -4.93 0.000 -.9196423 -.3963892 ---------------------+---------------------------------------------------------------- model_high_income_~r | _cons | 1.078982 .0489945 22.02 0.000 .9829543 1.175009 --------------------------------------------------------------------------------------
Thank you !
0 Response to see if regression coefficients are statistically different between 2 groups (SUEST)
Post a Comment