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

Re: st: interaction without main effects


From   John Hendrickx <[email protected]>
To   [email protected]
Subject   Re: st: interaction without main effects
Date   Tue, 14 Jan 2003 00:47:16 -0800 (PST)

Typing i.agegrp|durastay will create dummmies for the main effect of
agegrp and interactions between agegrp and durastay as a continuous
variable, but no main effects of durastay. But that isn't what the
original poster wanted.

You can use -desmat- for problems like this:

desmat: poisson patient origin age.durstay, exp(ptime)

This will estimate a model with main effects for origin and an
interaction between age and durstay, but no main effects of age and
durstay. But what you're perhaps really interested in is

desmat: poisson patient origin age durstay age3.durstay3, exp(ptime)

Where age3 and durstay3 are the variables recoded to 3 categories. 

The latest version of -desmat- is available at SSC. Type -ssc
describe desmat- for more information.

Hope this helps,
John Hendrickx

--- Jonathan Beck <[email protected]> wrote:
> typing i.agegrp|durastay (without "i."before durastay) should give 
> you main effects for durastay plus the interactions with agegrp,
> but 
> no main effects for agegroup.
> 
> I suppose you read the right thing in help, but got an error by 
> translating it to your varnames.
> 
> Jonathan
> 
> > Dear all,
> > I might just look at the wrong place, but...
> > I'm performing a poisson regression like:
> >  xi: poisson patient i.origin i.age i.durstay, exp(ptime) 
> > with 7 agegroups and 9 groups for durstay.
> > I want to add an interaction between age and durstay, defined as
> > i.agegrp*i.durastay, for which I reallocated agegroups and
> durstay
> > groups to broader categories 3*3. Now of course, the only extra
> output
> > I'd like to see is the interactions, not the main effects.
> > 
> > The only thing I could find in "help" is: 
> > i.varname1|varname3     creates dummies for categorical variable
> > varname1 and includes continuous variable varname3:  all
> interactions
> > and main effect of varname3, but not main effect of varname1.
> > 
> > So I conclude from that, that if I would do i.agegrp|i.durastay,
> I would
> > get the interactions and the main effect of durastay? That
> doesn't
> > really help, so what character should be used for only
> interaction and
> > no main effect? the FAQ's and websearch didn't really guide me to
> the
> > right direction too. 
> > I use stata 7
> > 
> > Thanks for any help! 
> > Annelies
> > -- 
> > A.M. Vos
> > dept. of Public Health
> > Erasmus MC
> > University Medical Center Rotterdam
> > P.O. Box 1738
> > 3000 DR Rotterdam
> > The Netherlands
> > Tel: +31 (10) 4087502
> > Fax: +31 (10) 4089449
> > email: [email protected]
> > *
> > *   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/
> 
> 
> *
> *   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/


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index