Richard Goldstein wrote:
I have estimated a poisson regression (includes some interaction terms
and uses cluster SE's). After estimation I obtained predicted values
(-predict newvar, ir-) and some of the IR's are greater than 1.0
(Similarly, using -predict newvar- gives some counts that are greater
than the exposure (or denominator).) I have not known this to occur
before and have the following questions:
1. is this a serious problem?
----------
No, it isn't a problem.
An incidence rate can take any non-negative value; it is the number of
events divided by the time at risk (the "exposure" in poisson regression
terms). So obviously the numerical magnitude of an incidence rate
depends on the time unit chosen (seconds, years, centuries).
Hope this helps
Svend
________________________________________________________
Svend Juul
Institut for Folkesundhed, Afdeling for Epidemiologi
(Institute of Public Health, Department of Epidemiology)
Vennelyst Boulevard 6
DK-8000 Aarhus C, Denmark
Phone, work: +45 8942 6090
Phone, home: +45 8693 7796
Fax: +45 8613 1580
E-mail: [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/
