I run multinomial probit and then run marginal effect for each outcome (Note. I have four outcomes for the dependent variable). In the past, I ran this way. The results of maginal effects came out. However, this time, I added a new variable "MEMBERS" which is a continous independent variable (which is the number of household members e.g., 1,2, 3, 4, 5,...14 etc.). Below are my syntaxes for each variable.
margins, dydx (AREA) at (AREA = 2) predict (pr outcome (1))
margins, dydx (Have_child0to14years) at (Have_child0to14years = 0) predict (pr outcome (1))
margins, dydx (Have_adult) at (Have_adult = 1) predict (pr outcome (1))
margins, dydx (Family_disrupt) at (Family_disrupt =2) predict (pr outcome (1))
margins, dydx (TypeHH_relationtohead) at (TypeHH_relationtohead = 2) predict (pr outcome (1))
margins, dydx (MEMBERS)predict (pr outcome (1))
My question is "Is there something wrong with my syntax for the variable "MEMBERS?" After running these syntexes (please see below), I got the marginal effect for the "MEMBERS" variable. But I did not get the marginal effect results for other variables *for example, below, there is no marginal effect, standard error, other values showing up. So what should I do to get marginal effect results?
Average marginal effects Number of obs =
> 8102
Model VCE : OIM
Expression : Pr(Work_studycross==1), predict(pr outcome (1))
dy/dx w.r.t. : 1.TypeHH_relationtohead 3.TypeHH_relationtohead
4.TypeHH_relationtohead 5.TypeHH_relationtohead
8.TypeHH_relationtohead
at : TypeHH_relationtohead= 2
--------------------------------------------------------------------
> -------------------
| Delta-method
| dy/dx Std. Err. z P>|z| [
> 95% Con
> f. Interval]
----------------------+---------------------------------------------
> -------------------
TypeHH_relationtohead |
1 | . (not estimable)
2 | 0 (empty)
3 | . (not estimable)
4 | . (not estimable)
5 | . (not estimable)
8 | . (not estimable)
--------------------------------------------------------------------
> -------------------
Note: dy/dx for factor levels is the discrete change from the base level.
. margins, dydx (MEMBERS)predict (pr outcome (1))
Average marginal effects Number of obs =
> 8102
Model VCE : OIM
Expression : Pr(Work_studycross==1), predict(pr outcome (1))
dy/dx w.r.t. : MEMBERS
--------------------------------------------------------------------
> ----------
| Delta-method
| dy/dx Std. Err. z P>|z| [95% Conf.
> Interval]
-------------+------------------------------------------------------
> ----------
MEMBERS | -.0020091 .0019131 -1.05 0.294 -.0057587
> .0017405
--------------------------------------------------------------------
> ----------
Many thaniks & regards,
Kanoknit
Thailand
Comment