RE: st: Tobit interaction effect

[Ebru Ozturk <ebru_0512@hotmail.com>]

Thank you very much Maarten. But, what do you mean with other interpretation? Could you give an example?
 
Ebru


----------------------------------------
> Date: Thu, 10 May 2012 11:17:11 +0200
> Subject: Re: st: Tobit interaction effect
> From: maartenlbuis@gmail.com
> To: statalist@hsphsun2.harvard.edu
>
> On Wed, May 9, 2012 at 10:31 PM, Ebru Ozturk wrote:
> > I would like to measure the interaction effects as it is in OLS. My data is cross-sectional and dependent variable is censored. So, I guess I need to see the expected value of the observed value. I use Stata 10.
>
> Not necessarily. Given that information I could just as easily justify
> the interpretation in terms of the latent variable or the observed
> values and the censored values fixed at the cut-off point.
>
> As I said before: if you interpret your results in terms of the latent
> variable you can just do what you would always do in a linear
> regression(*). When you want the other interpretation you will need to
> derive the cross partial derivatives yourself. Once you have done that
> you can use -predictnl- to compute those effects and their standard
> error for each observation.
>
> -- Maarten
>
> (*) Notice that OLS is just an algorithm while linear regression is
> the model. I guess you mean the latter and not the former.
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
>
> http://www.maartenbuis.nl
> --------------------------
> *
> * 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/ 		 	   		  
*
*   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/