Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Interaction after XTPROBIT, RE


From   Fra GMAIL <[email protected]>
To   [email protected]
Subject   st: Interaction after XTPROBIT, RE
Date   Thu, 10 May 2012 12:25:11 +0200

> 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: [email protected] or
[email protected] or  [email protected]
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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index