Re: st: Binary Choice Model and fixed effects - interpreting the interaction effects?

[Benjamin Niug <benjamin.niug@googlemail.com>]

@Maarten. Thanks. I tried to calculated the marginal effects as
indicated in the paper you mentioned (M.L. Buis (2010) "Stata tip 87:
Interpretation of interactions in non-linear models", The Stata
Journal, 10(2), pp. 305-308)

However, some interactions are not estimated / "estimable" by Stata
using the -margins- command.

In detail:

-margins, over(ABC below_median_expenses) expression(exp(predict(xb)))	post-

Stata output:
----------------------------------------------------------------------------------------------------------

Predictive margins                                Number of obs   =       3348
Model VCE    : robust

Expression   : exp(predict(xb))
over         : ABC below_median_expenses

Delta-method
Margin   Std. Err.      z    P>z     [95% Conf. Interval]
ABC#below_medi~s

0 0    (not estimable)
0 1     .2754675   .0348275     7.91   0.000     .2072069    .3437281
1 0    (not estimable)
1 1     .3525662   .0636748     5.54   0.000     .2277659    .4773666
----------------------------------------------------------------------------------------------------------

fyi: the regression command was:

- clogit labor_participation  ABC below_median_expenses
ABC_below_median_expenses , group(country_psu) vce(cluster
country_xxx) or -

where ABC_below_median_expense indicates the interaction effect.

Notice: there is a sufficient number of observations for each
interaction (> 200) - I have just checked that.

Best,
Benjamin
*
*   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/