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

[Nick Cox <njcoxstata@gmail.com>]

-inteff- is from SJ.

SJ-4-3  st0063_1  . . . . . . . . . . . . . . . . . Software update for inteff
        (help inteff if installed)  . . . . . E. C. Norton, H. Wang, and C. Ai
        Q3/04   SJ 4(3):359
        bug fix for inteff

SJ-4-2  st0063  Interaction effects & standard errors in logit & probit models
        (help inteff if installed)  . . . . . E. C. Norton, H. Wang, and C. Ai
        Q2/04   SJ 4(2):154--167
        computes the marginal effect of a change in two interacted
        variables for a logit or probit model, provides standard
        errors, and graphs the interaction effect

In referring to user-written commands, you are asked to make clear
_where_ the command comes from. -inteff- only supports -logit- or
-probit-. I don't know where you got the impression that either
-xtlogit- or -clogit- is supported. I can't comment on how to get what
you want or whether it is a good idea.

Nick

On Mon, Apr 2, 2012 at 10:57 AM, Benjamin Niug
<benjamin.niug@googlemail.com> wrote:

> I want to estimate a binary choice model accunting for time-invariant
> fixed effects (I read I could use the -xtlogit- or -clogit- command).
>
> y_it = b_1*x_1_it*x_2_it+b_2*x_1_it + b_3*x_2_it
>
> However, I have included an interaction effect which I want to
> interpret correctly - as pointed out by Ai and Norton (2004) this is
> not trivial. They suggest to use a user written command called
> -inteff-. This command works well if -logit- is used, however, it does
> not work if -xtlogit- or -clogit- is used.
>
> Error:
> "last estimates not found
> r(301);"
>
> My Stata code:
>
> clogit labor_participation ABC minority ABC_minority expenses  Obvious
> ageHH HH_female profess informal_employed language muslim,
> group(country_xxx) vce(cluster country_xxx)
>
> inteff labor_participation  ABC minority ABC_minority expenses
> Obvious ageHH HH_female profess informal_employed language muslim,
> savedata(ABC_minority_rural, replace) savegraph1(ABC_minority_rural_1,
> replace) savegraph2(ABC_minority_rural_2, replace)
>
> Does anybody have an idea why this is the case and how I can still get
> the correct interpretations of the interaction effects?

*
*   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/