st: Interaction after XTPROBIT, RE

[Fra GMAIL <francescodilorenzo80@gmail.com>]

> Hi everyone,
>
> I'm estimating the likelihood of individual inter-firm mobility with 2 main continuous explanatory vars: X1 (only positive values) and X2 (only negative values) and some control vars. I use the command -xtprobit, re-, I would like to interact both Xs with the following vars:
>
> Z1: discrete var
> Z2: continuous var
> Z3: dummy
>
> In my full model, should I have all of them together (6 interaction terms)? Or, should I run as many full models as the # of Zs?
> In the case I run each of them separately, I still have 2 interaction terms in each model (b1X1 + b2X2 + b3Zi + b4X1Zi + b5X2Zi).

> Following Norton et al. (2004) -Inteff- command allows only for 1 interaction in each model, correct? So, it seems I can't use it (unless I run as many models as each single interactions, but would it be correct?). So, how can I do this?

>In addition, using -inteff- after -xtprobit, re- apparently does not work? Is -inteff-  a command only for non-panel data structure? So, in my case, how do I solve this?
>
> Many thanks
>
> Francesco
>
>
>
>



--
Francesco Di Lorenzo
Address: Avinguda Republica Argentina, 10, 5-1, 08023, Barcelona, Spain
Mobile: +34 697 45 05 61 (ESP) +39 333 27 95 429 (ITA) or +1 202 422 2726 (USA)
Emails: francesco.dilorenzo@esade.edu or
francescodilorenzo80@gmail.com or  fd75@georgetown.edu
Website: https://sites.google.com/site/dilorenzof2011/
Skype: trontositu

*
*   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/