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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Using # sign with factor variables to create interactions


From   Christopher Swearingen <christopher.swearingen@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Using # sign with factor variables to create interactions
Date   Fri, 6 Aug 2010 17:02:56 -0400

The factor variable interactions are available for post-estimation using LINCOM.

e.g. ---

logistic outcome i.smoker##i.agegrp

lincom 1.smoker + 2.agegrp + 1.smoker#2.agegrp

You should be able to estimate all levels in this manner.

Best regards,

~c
________________________________________________________

Christopher Swearingen, Ph.D.
Assistant Professor of Pediatrics and Biostatistics
University of Arkansas for Medical Sciences
Department of Pedatrics
One Children's Way, Slot 512-43
Little Rock, AR 72202

http://www.arpediatrics.org/research/biostatistics



>
> Hi Statalisters,
>
> I'm using the new # sign notation with factor variables to create
> interactions.  I would like to do post model linear combinations with
> the lincom command.  Since the level indicator variables are not created
> in the dataset (or memory I believe), how do I do linear combinations
> after running a model that incorporated an interaction using the # sign
> notation (ex. smoker#agegrp)?  I've discovered that I could just change
> the base level in order to see different linear combinations but this
> requires running new models for each combination I want.  Is this the
> best way to do it with the new notation?
>
> Tom Weichle

*
*   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