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

st: cluster


From   dengue <[email protected]>
To   [email protected]
Subject   st: cluster
Date   Wed, 27 Jul 2005 17:39:38 +0900 (JST)

Dear all,

I am analysing three datasets, each of which
are from the same population bu different years, 
say 2002, 2003 and 2004, but not longitudinally
coded;

The number of cases of a disease (numb_of_cases),
the population size (popul)
the degree of exposure (exposure)
are recorded in each of 1000 towns. 

The incidence (number of cases per
population) is about 1/1000 per year.

Initially, I used such commands as
.glm numb_of_cases exposure if year==2002, f(b popul)
.glm numb_of_cases exposure if year==2003, f(b popul)
.glm numb_of_cases exposure if year==2004, f(b popul)

Q1. I would like to bundle these three years
into one analysis.  "cluster" option is
available for such a command as "regress",
but not for "glm".  How could this analysis
be clustered?

Q2.  I was suggested that
familly binomial may not be adequate;
logistic or poisson regression may be more
suitable.  What commands can be used for
the above settings?

Let me apologise to you for these lay
questions.

Nagao 

__________________________________
Save the earth
http://pr.mail.yahoo.co.jp/ondanka/

*
*   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