marginal effect after two-stage multinomial logit

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

marginal effect after two-stage multinomial logit

This post has NOT been accepted by the mailing list yet.
Dear all,

I want to estimate a following two-stage multinomial logit model (the dependent variable has 3 categories) with an endogenous variable, and get the marginal effects. However, the following program doesn't work. I always get some error message as "insufficient observations to compute bootstrap standard errors
no results will be saved ".

Could you please check the problems for me ? Thank you very much!

The following is my program:

webuse sysdsn1, clear
program myprog
version 11.1

//stage 1
regress ppd  male nonwhite age
predict double ppdhat, xb

//stage 2
quietly mlogit insure male nonwhite ppdhat
margins, dydx(*) predict(outcome(1))
margins, dydx(*) predict(outcome(2))
margins, dydx(*) predict(outcome(3))


bs, reps(1000) seed(1014): myprog