Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: plotting interaction in Tobit analysis


From   frone@ria.buffalo.edu
To   statalist@hsphsun2.harvard.edu
Subject   st: plotting interaction in Tobit analysis
Date   Wed, 4 Jun 2003 15:34:06 -0400

I'm aware of how to plot an interaction from linear regression, but I'm 
having some trouble figuring out how to plot an interaction from a Tobit 
regression where the predicted values are constrained to fall within the 
range of the observed outcome.

For example, my variables are:

alc=alcohol use
race=race (0=white, 1=minority)
pa=postive affect (scale scores range from 1 - 4)
int=race*pa

My Tobit results are:

alc = 1.37 + .42(pa) - 5.89(race) + 1.84(int)

One can rewrite this equation to get the separate equations for white and 
minority:

white: alc = 1.37 + .42(pa)

minority: alc = -4.52 + 2.26(pa)

Now consider the equation for minorities, if I compute the predicted value 
for alc at the lowest level of pa, I get

minority: alc = -4.52 + 2.26(1) = -2.26

Of course, the lower limit of the observed variable is zero unlike the 
latent outcome. 

Is there a way to plot this interaction where the predicted alcohol values 
fall within the range of the observed outcome?

Thanks in advance for any help.

Mike Frone


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index