Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

RE: st: biprobit predict command


From   "Garrido, Melissa" <melissa.garrido@mssm.edu>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: biprobit predict command
Date   Thu, 26 May 2011 14:26:15 +0000

Hi Jean-Christophe,
One thing to consider with your first question is the issue of interpretability.  What does a prediction based on the mean of a binary variable represent? For example, if your binary variable was whether someone was married, you'd be estimating the effect of hybrid individuals who were/were not married. Depending on how many binary IVs you have, it may be worth considering holding individual IVs constant at 0 or 1 and then getting the predictions.
I'm not sure how to answer your second question; hopefully someone else has an answer for you.

- Melissa


>From 	  Jean-Christophe Boucher <Jean-Christophe.Boucher@hei.ulaval.ca>
>To 	  "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
>Subject 	  st: biprobit predict command
>Date 	  Wed, 25 May 2011 16:21:30 -0400

>Hi,
>I am using a bivariate probit model with partial observability (biprobit)
>where some of the IVs in the first equation are either binary or continuous.
>Right now I have used the adjust stata command in order to predict
>estimations of the continuous IVs while keeping all other IVs at their
>means. There are two issues that worries me:
>1) Is keeping binary IVs at their means to estimate predictions is worrisome
>2) In bivariate models, the prediction commands only works for the first
>equations, should I be worried about not accounting for the rho in the
>prediction? Or is it already included in the biprobit results? Is there any
>specific command to use post-estimation technics for multilevel model such
>as the one I am using?

>JC
-------------------------------------------------------------------- 
Melissa Garrido, PhD 

Research Health Science Specialist 
GRECC/REAP, James J Peters VA Medical Center 
130 West Kingsbridge Road 
Bronx, NY 10468 
718-584-9000 x 3804 
melissa.garrido@va.gov 

Assistant Professor 
Brookdale Department of Geriatrics & Palliative Medicine 
Box 1070 
Mount Sinai School of Medicine 
One Gustave L. Levy Place 
New York, NY 10029 
melissa.garrido@mssm.edu 


*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index