-mfx compute- will compute the marginal effects after -xtlogit, fe- with the
predict(pu0) option. The pu0 comes from -clogit- which also estimates
conditional fixed-effects logit
models.
I am not sure why "no effect" is reported when using the -predict(p)- option.
. webuse union
(NLS Women 14-24 in 1968)
. qui xtlogit union age grade south year, fe i(id)
. mfx compute, predict(pu0) nose
Marginal effects after clogit
y = Pr(union|fixed effect is 0) (predict, pu0)
= .16861097
-------------------------------------------------------------------------------
variable | dy/dx X
---------------------------------+---------------------------------------------
age | .0106352 30.538
grade | .0120169 12.7934
south*| -.099099 .381388
year | -.0083177 79.6184
-------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1
. mfx compute, predict(p) nose
Marginal effects after clogit
y = Pr(union|single outcome w/i idcode) (predict, p)
= 1
-------------------------------------------------------------------------------
variable | dy/dx X
---------------------------------+---------------------------------------------
age | (no effect) 30.538
grade | (no effect) 12.7934
south*| (no effect) .381388
year | (no effect) 79.6184
-------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1
Hope this helps,
Scott
----- Original Message -----
From: "Bo MacInnis" <[email protected]>
To: <[email protected]>
Sent: Tuesday, March 02, 2004 6:54 PM
Subject: st: How to calculate marginal effects after xtlogit, fe?
> Hi,
>
> Would anyone help me to calculate the marginal effect of a binary regressor
> after I estimate a fixed effect conditional logit? "mfx compute" gives me
> something that I am afraid is not what I am looking for.
>
> Thank you so much,
> Bo MacInnis
*
* 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/
