I think your best bet might be to use poisson:
Create indicators for trainers using
. tabulate trainer, generate(X)
Use poisson to get rates by trainer
. poisson fract X* , e(time) nocons
where fract and time are my guesses at your variable names
Use testparm to test for differences in rates
. testparm X*, equal
Use poisson as above but without the X* to get an overall rate.
On Fri, Jun 06, 2003 at 11:47:43AM +0100, KRISTIEN VERHEYEN wrote:
> Dear all,
>
> I have dataset from a cohort study including horses, trainers, number of
> fractures and days at risk. Each observation consists of a horse with its ID
> number, the trainer by which it was trained, the number of fractures it
> sustained during the study period and the number of days it contributed to
> the study.
> >From this, I would like to produce a table showing fracture incidence rates
> by trainer and also a weighted overall incidence rate, giving more weight to
> trainers who contributed more days at risk to the study. Also, is there a
> way Stata can tell me if the incidence rates between the trainers vary
> significantly (without having to produce IRR)?
> There are 13 trainers in the dataset, training 1178 horses, providing
> 386,803 days at risk.
>
> At this stage, I do not wish to stset my data, as there are some
> complications there with gaps and repeat events and I wouldn't think it 'd
> be that complicated to produce some simple IR and comparisons. I have used
> Stata help, searched FAQ and browsed the books but can not seem to find
> answers to these rather simple questions. Thanks very much for any help.
>
> Kristien Verheyen
>
> **********************************************
> Visit our website at www.aht.org.uk
> **********************************************
>
> #####################################################################################
> This e-mail message has been scanned for Viruses and Content and cleared
> by MailMarshal
> For more information please visit www.marshalsoftware.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/
--
Michael
[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/
