Re: st: marginal effects after multinomial logits

[John Francis <francisj71@googlemail.com>]

Hi Luciana,

you should try -estout-.
first, you run your regression, like
.mlogit y x
.est sto m

Then, you get the marginal effects (I use four categories of the
dependent variable in this example)
.forval i = 1/4 {
.est res m
.margins, dydx pr(out(`i')) post
.est sto m`i'
.}

Finally, you use -estout- to create your table
.estout m1 m2 m3 m4, cells(b se)

This is a very simple example, you should get acquainted with estouts
options to get the table you want.

John


On Thu, Mar 22, 2012 at 10:08 AM, Luciana Mèndez <lucimerrico@gmail.com> wrote:
> Hy!
>
> I am running a standar multinomial logit regression,
> after that I run margins (i.e.: margins, dydx(*) predict (pr
> outcome(2))) to get the mean marginal effect, I would like to get the
> marginal effects of eahc outcome of my dependent variable into a nive
> table, however outreg2 wouuld only produce an output tabl based on the
> mlogit coefficients,
> I tried both:
>
> outreg2 using XXX: margins, dydxx(*)
>
> as well as
>
> margins, dydx(*),
> outreg2 using XXX
>
> I there ary other way to get the margins results directly in a
> pblication- type table?
> thank you!
> Luciana
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/