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

RE: st: plot predicted effects after regression


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: plot predicted effects after regression
Date   Sat, 8 May 2004 18:58:14 +0100

I would be surprised if that were true. 

However, it seems to me that the challenge is 
not in the graphing; it is in the calculation. 

You can use -adjust-: you just need to talk  
your way past the requirement for a -by()- 
option (unless that is part of what you want). 

Here is a silly example: 

. sysuse auto 
. regress mpg weight headroom turn trunk length displacement
. gen all = 1 
. adjust headroom turn trunk length displacement, by(all) gen(predict) 
. scatter predict weight

yumin sheng
 
>   Thanks so much. Your solution is great, but if I
> remember correctly, I think Stata has a ready and very
> simple command for post-estimation graphing of the
> predicted effects. 

Thomas Trikalinos 

> > So you need predicted values on the
> > VariableOfInterest adjusting at the 
> > mean level of continuous covariates, and the
> > reference category of 
> > categoric covariates.
> > 
> > A simple but not so elegant  solution is
> > 
> > . gen PredY = Constant + beta1* VariableOfInterest1
> > + 
> > beta2*MeanContinuousCovariate2   (or the
> > corresponding analogue for a 
> > logit/probit/poisson etc regression)
> > 
> > [first run . egen MeanContinuousCovariate2 =
> > mean(ContinuousCovariate2)]
> > 
> > Constant and beta1, beta2 are from the regression
> > output. All 
> > Categorical covariate terms are zero (this would be
> > your reference 
> > category, right?) and all the continuous covariate
> > terms are 
> > incorporated using their mean level. This way you
> > adjust for the 
> > reference category for categoric covariates and for
> > the mean value of 
> > continuous covariates.
> > 
> > You most probably have more than one continuous
> > covariates; just put in 
> > as many terms as you need. If you have different
> > functions of the 
> > VariableOfInterest (eg quadratic or cubic terms) put
> > them in as more 
> > VariablesOfInterest.
> > 
> > This is a crude workaround I use. I'm confident that
> > people know 
> > something better and more elegant, though...

*
*   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