After -mfx compute- you don't get an error? Try the pu0 predict option.
. sysuse auto
. qui xtlogit fore mpg, i(rep) fe
. mfx
predict() expression unsuitable for marginal-effect calculation
r(119);
. mfx, predict(pu0)
Marginal effects after xtlogit
y = Pr(foreign|fixed effect is 0) (predict, pu0)
= .9335091
----------------------------------------------------------------------------
variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X
---------+------------------------------------------------------------------
mpg | .0075941 .00548 1.39 0.166 -.003148 .018336
21.5932
----------------------------------------------------------------------------
Scott
> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of Arkes, Jeremy
> Sent: Tuesday, October 04, 2005 11:58 AM
> To: [email protected]
> Subject: st: marginal effects in fixed-effect logit model
>
> Hi,
>
> I've been trying to calculate marginal effects in a fixed-effect logit
> model, with the XTLOGIT command. My model is the following:
>
> xtlogit y x1 x2 x3 x4, i(hhid) fe
>
> The model runs fine and has 2 of the X variables being significant. When
> I do "mfx compute" after the model, the output indicates "no effect" for
> each X variable.
>
> Is there something I might be missing?
>
> Thanks,
>
> Jeremy
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
