Hurdle logit-Poisson model: marginal effects equal the coefficient estimate

[Kerstin Schmidt]

Dear Statalist,

Having estimated the coefficients in a Hurdle Logit-Poisson model by:

svy: hplogit y $xvars

and then the marginal effects for both stages:

margins, dydx(*) predict(eq(logit))
margins, dydx(*) predict(eq(poisson))

produces the same results for coefficient and marginal effect. This is also the case in the example posted by Cameron, Adrian Colin, and Pravin K. Trivedi. Microeconometrics using stata. Vol. 5. College Station, TX: Stata press, 2009. on page 572 and 574.

What does mean? Are coefficient and marginal effect the same or is there is a mistake in the commands?

Any help is appreciated!

[Jessica Weaver]

Hi Kerstin,
Did you ever find an answer to this? I am having a similar issue. I'd love to hear if you figured it out.

Thanks!

[Richard Williams]

Confusion over things like this usually stems from not understanding what the default option for margins is. You are probably getting xb when you want pr (for the logit part anyway). If you look closely at the margins output you will see what it is giving you. Others can correct me if I am wrong, but I think the correct coding for the logit part is more like

use https://www3.nd.edu/~rwilliam/statafiles/couart4.dta, clear
hplogit art female married kid5 phd mentor, nolog robust
margins, dydx(*) predict(eq(logit))
margins, dydx(*) expression(invlogit(predict(eq(logit))))

[Teresa Mungazi]

How about for the count part? What is the syntax?

[Andrew Musau]

See https://www.statalist.org/forums/for...=1720219355388.