Interpreting interaction effects in non-linear models can be tricky. See the following article and related Stata code called inteff.
________________________________
From: [email protected] on behalf of Ashwin Ananthakrishnan
Sent: Wed 7/20/2005 11:57 PM
To: [email protected]
Cc: [email protected]
Subject: st: Interaction terms in logistic regression using dummy variables
Hi,
I'm having some trouble interpreting the interaction
terms in logistic regression using stata.
My two variables of interest are gender (0-male,
1-female), and age categories (0- 65-69, 1-70-74,
2-75-79, 4-80 and above), and my outcome variable is
anyscreen.
When i type in the command,
xi: logistic anyscreen i.gender*i.income, I get the
following terms
-Igender_1, _Iagecat_1, _Iagecat_2, _Iagecat_3,
_IgenXage_~1,_IgenXage_~2, _IgenXage_~3.
What do teh odds ratios for each of these terms mean -
especially the interaction terms. And how do I test
the significance of these interaction terms.
Thanks,
Ashwin
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.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/
<<winmail.dat>>
