@NickWolfinger @Stata First off, you should try using the i. in place of the c. operator for year. The latter treats year as a continuous variable and assumes that the coefficients of the relationships remain constant across time. The former allows them to vary by year.
@NickWolfinger @Stata So it should be educ##year as opposed to educ##c.year (no need to include the i. operator).
@NickWolfinger @Stata With i.year, you also wouldn't have to manually specify at(year=(1980 1981 1982) etc. etc.
So if you'd like to see how the effects of having a college education on Y change over time, you would do...
margins i.year, at(educ=(insert value for college education)).
So if you'd like to see how the effects of having a college education on Y change over time, you would do...
margins i.year, at(educ=(insert value for college education)).
@NickWolfinger @Stata and if you'd like year-conditional estimates for *all* levels of education, you would just do... margins educ#year.
Not sure whether any of this answers your question, but thought I'd try.
Not sure whether any of this answers your question, but thought I'd try.
