On Monday, Gang wrote:
> Thanks for the update. But the point is I need to know how to use it.
>
> In logit, I run mfx after logit can get the marginal effect at the mean
> value of the all predictors; how can I do this under xtlogit? Could you
> help?
The predict option -pu0- after -xtlogit, fe- is often used, as in
this example:
clear
sysuse auto
xtlogit for mpg, i(head) fe
mfx, predict(pu0)
Let me stress, that there is not one correct predict option and all
others are incorrect. A marginal effect is a derivative of a function of
the coefficients of the model. So we may calculate different marginal
effects by differentiating different functions.
For example, using the predict option -xb- with -mfx- after -xtlogit,
fe- gives a sensible answer, but it probably isn't the one we want. So
which predict option we use depends on which marginal effect we are
after.
--May
[email protected]
*
* 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/
