Hello stata friends,
So I am currently estimating a multinomial logit model with random effects. I have been able to calculate the results but I am struggling to calculate the marginal effects.
My code is the following:
But I get an error that states "prediction is a function of possibly stochastic quantities other than e(b)"
Does anybody have some insight?
Thank you!
So I am currently estimating a multinomial logit model with random effects. I have been able to calculate the results but I am struggling to calculate the marginal effects.
My code is the following:
Code:
gsem (i.mk_choice_ <- edad_ sexo_jh o2_2_d1_ o2_2_d4_ o2_2_d7_ hato_ hato2_ /* */perc_propia_ index_assets_ altitud promed_prec index_viol_base i.clasificacion big_region med_region /* */year_2013 educ_none RI0[folio]@1), mlogit margins , dydx(_all) atmean predict(outcome(1.mk_choice_ ))
But I get an error that states "prediction is a function of possibly stochastic quantities other than e(b)"
Does anybody have some insight?
Thank you!