Hello everybody,
After running a FE model, I am trying to obtain the marginal effects of a continuous variable "prop_lessthan_HSedu" differentiated by a categorical variable "rooted_partisanship". Both variables are time-invariant. Morevoer, I included them in a three-way interaction with quarters "qdate". Here is the issue, the marginsplot displays fitted values instead of linear predictions. I, therefore, have two questions:
1. How to interpret the graph;
2. How I can get marginal effects for the continuous variable prop_lessthan_HSedu differentiated by rooted_partisanship.
I spent quite a while on this, and I struggle to find an answer to this problem. Thank you in advance to whoever is willing to help, it would be appreciated!
After running a FE model, I am trying to obtain the marginal effects of a continuous variable "prop_lessthan_HSedu" differentiated by a categorical variable "rooted_partisanship". Both variables are time-invariant. Morevoer, I included them in a three-way interaction with quarters "qdate". Here is the issue, the marginsplot displays fitted values instead of linear predictions. I, therefore, have two questions:
1. How to interpret the graph;
2. How I can get marginal effects for the continuous variable prop_lessthan_HSedu differentiated by rooted_partisanship.
I spent quite a while on this, and I struggle to find an answer to this problem. Thank you in advance to whoever is willing to help, it would be appreciated!
Code:
asdoc xtscc series_complete_pop_pct i.rooted_partisanship##i.qdate##c.prop_lessthan_HSedu /// L7.(total_cases_rate total_deaths_rate) i.mdate, fe lag(2) replace margins rooted_partisanship, at(prop_lessthan_HSedu=(0(15)75)) vsquish margins, noci
Attached Files
