st: Re: RE: RE: mfx after xtnbreg and how to compute predicted Y

["Martin Weiss" <martin.weiss1@gmx.de>]

The additional x-column gives the means of the covariates at which the 
marginal effect is estimated.

*********
webuse airacc, clear
xtnbreg i_cnt inprog, exposure(pmiles) irr
mfx
*let´s see the mean of "inprog" from the "x-column"
su inprog,mean
di in red r(mean)
*********

-mfx evaluates at the mean as default. See -help mfx##atlist-

HTH
Martin

----- Original Message ----- 
From: "Pek-Hooi Soh" <phsoh22@gmail.com>
To: <statalist@hsphsun2.harvard.edu>
Sent: Sunday, October 19, 2008 8:47 PM
Subject: st: RE: RE: mfx after xtnbreg and how to compute predicted Y


> Hi Nick,
>
> I was told that the coefficients of non linear models cannot be used
> directly for interpretation, and I should use mfx that give the marginal
> effects (exp(B),I think). In the table, the dydx column output is the same
> as the coefficients of xtnbreg, the x column is the only extra 
> information.
> What is x?
>
> Cheers,
>
> Pek
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
> Sent: October-19-08 11:21 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: mfx after xtnbreg and how to compute predicted Y
>
> -mfx- is giving you extra information for the model you just ran.
>
> The help for -xtnbreg postestimation- indicate how to get predictions in
> Stata. There is no need to resort to any other software.
>
> Nick
> n.j.cox@durham.ac.uk
>
> Pek-Hooi Soh
>
> I have a count panel data model with interaction terms where the terms
> are mean centered.
>
> I estimated the full model by -xtnbreg- and ran -mfx-. The marginal effect
> table presents the same coefficients and standard errors as the table for
> -xtnbreg-. Is this possible?
>
> If I do get the correct -mfx- output, how do I compute predicted Y? Do I
> have
> compute y = exp(XB) where B stores marginal effects separately in Excel ?
>
> *
> *   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/

*
*   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/