Hi,
I am working with panel data, and I am using two types of regressors: one is own illness, and the other is a set of 6 "other-age-sex-ill" variables, which are binary (equal to 1 if at least one other household member in the same age-sex category is ill, and 0 otherwise). and the following are the command:
reghdfe individual consumption i.illness##i.age_cat other_child_male_ill other_child_female_ill other_adult_male_ill other_adult_female_ill other_elderly_male_ill other_elderly_female_ill [individual specific control] , absorb(indv hh#year) cluster(village)
So, I have a doubt if I want to see the effect of other members' illness (in the specific age-sex category) on one's own consumption. Is the creation of other_child_male_ill... and so on the right way, and does including household-year fixed effects make this variable collinear with it?
I am working with panel data, and I am using two types of regressors: one is own illness, and the other is a set of 6 "other-age-sex-ill" variables, which are binary (equal to 1 if at least one other household member in the same age-sex category is ill, and 0 otherwise). and the following are the command:
reghdfe individual consumption i.illness##i.age_cat other_child_male_ill other_child_female_ill other_adult_male_ill other_adult_female_ill other_elderly_male_ill other_elderly_female_ill [individual specific control] , absorb(indv hh#year) cluster(village)
So, I have a doubt if I want to see the effect of other members' illness (in the specific age-sex category) on one's own consumption. Is the creation of other_child_male_ill... and so on the right way, and does including household-year fixed effects make this variable collinear with it?
Comment