Hello Statalist!
I am trying to calculate the marginal effects of an ordered logit. I have read some advice that calculating marginal effects at fixed values of continuous covariates is sometimes a more digestible/useful way of using marginal effects than calculating average marginal effects.
I am trying to calculate the marginal effect of one of my explanatory variables, gender inequality (an index bound between 0 and 1) on my dependent variable reported happiness (ordinal categorical variable, takes values 1,2,3 or 4). Calculating the marginal effect has been fine. However, when I try to fix the values of gender inequality, stata tells me it is 'not estimable'. Would anyone be able to help me understand why?
My ordered logit regression is:
The code stata will not estimate is:
Stata has no problem with just
Thank you very much in advance!
I am trying to calculate the marginal effects of an ordered logit. I have read some advice that calculating marginal effects at fixed values of continuous covariates is sometimes a more digestible/useful way of using marginal effects than calculating average marginal effects.
I am trying to calculate the marginal effect of one of my explanatory variables, gender inequality (an index bound between 0 and 1) on my dependent variable reported happiness (ordinal categorical variable, takes values 1,2,3 or 4). Calculating the marginal effect has been fine. However, when I try to fix the values of gender inequality, stata tells me it is 'not estimable'. Would anyone be able to help me understand why?
My ordered logit regression is:
Code:
ologit Q46_new c.genderinequality job_scare election_equality home_equality political_equality_perception GDPpercap1 giniWB unemploytotal i.ISO31661numericcode $X, vce(cluster ISO31661numericcode)
Code:
margins, dydx(genderinequality) at(genderinequality=(0.3 0.6))
Code:
margins, dydx(genderinequality)
