Re: st: Regression with multiple age groups

[brendan.halpin@ul.ie (Brendan Halpin)]

On Wed, May 02 2012, Shirley Sy wrote:

> Hi David,Thanks for your replies, which have been very helpful! The
> stage I am stuck on is dealing with the number of categorical variables
> I have. For example, say all I wanted to do was run a simple poisson
> regression with the data I have, how would I go about doing so when I
> have multiple subgroups for certain variables (i.e. husband's/wife's
> previous marital status, husband's/wife's age at divorce, duration of
> marriage) that contain count data of the total number of divorces in
> that year (spanning over 20 years) rather than individual level
> data?I've looked online extensively but can only find examples of
> individual level data or examples which are only applicable to one year
> and one categorical variable. 

I think it may be helpful to think in terms of loglinear models (which
are poisson regression underneath the hood). As a conceptual framework,
though, they help you think about modelling this sort of data,
particularly where you don't have individual data but numbers in cells
of cross-classifications.  

A quick google unearths this ref, which is promising in that it works
with vital statistics data, and models rates:
Log-linear Rate Modeling of Cross-classified Divorce Data
Abada, Teresa S. J.
January 2000
Canadian Studies in Population;2000, Vol. 27 Issue 1, p209

Brendan
-- 
Brendan Halpin,   Department of Sociology,   University of Limerick,   Ireland
Tel: w +353-61-213147  f +353-61-202569  h +353-61-338562;  Room F1-009 x 3147
mailto:brendan.halpin@ul.ie    ULSociology on Facebook: http://on.fb.me/fjIK9t
http://teaching.sociology.ul.ie/bhalpin/wordpress         twitter:@ULSociology
*
*   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/