Hello,
I´m working with a panel dataset, where each row corresponds to a county-week. I am modeling a fixed effects Poisson regression, with a categorical independent variable (4 types of treatment - type of school instruction on each week) and a count outcome (count of reports by week):
xtset county week
xtpoisson num_rpts_county i.instruction_county, exposure(population) vce(robust) fe
I would like to incorporate a count mediator but I'm not sure if it's possible to adapt the mediate command (for a count outcome and a count mediator) to a panel dataset and a fixed effects model. I found this sem discussion on mediation in panel data with a continuous outcome but I'm not sure whether/how this can be adapted to a count outcome.
Any insight/advice on this issue would be greatly appreciated!
I´m working with a panel dataset, where each row corresponds to a county-week. I am modeling a fixed effects Poisson regression, with a categorical independent variable (4 types of treatment - type of school instruction on each week) and a count outcome (count of reports by week):
xtset county week
xtpoisson num_rpts_county i.instruction_county, exposure(population) vce(robust) fe
I would like to incorporate a count mediator but I'm not sure if it's possible to adapt the mediate command (for a count outcome and a count mediator) to a panel dataset and a fixed effects model. I found this sem discussion on mediation in panel data with a continuous outcome but I'm not sure whether/how this can be adapted to a count outcome.
Any insight/advice on this issue would be greatly appreciated!
Comment