The previous cflpois had an error when the "by" variable was without a label.
The procedure has been updated to v1.02 works in Stata 6 & 7.
Example of call and output is (artificial data):
. cflpois count riskdays, by(period) dec(2) mult(100) n(persons)
period N count riskdays rate l a h
3 1325 349 56599 0.62 0.55 - 0.68
5 1400 374 68160 0.55 0.49 - 0.61
8 1435 381 78721 0.48 0.44 - 0.54
10 1569 414 89329 0.46 0.42 - 0.51
.... etc for all values of "per"
46 2896 459 299579 0.15 0.14 - 0.17
t N count riskdays rate l a h
Total 43008 8711 3631727 0.24 0.23 - 0.24
The help file has been modified and a new option added "dec( )" which
allows the user to determine number of decimals.
All cflpois does is actually to collapse data based on the variable (-les)
indicated in the by option and then passes the actual counts and time at
risk to cii procedure. Therefore the reference for calculations is now
Stata procedure cii.
Before it was: Haenszel W, Loveland D, Sirken MG. Lung cancer mortality as
related to residence and smoking histories. J Natl Cancer Inst 1962; 28:
947-1001
reprinted in: Breslow NE, Day NE: Statistical Methods in cancer Research.
Lyon,
IARC, 1987. P. 70.
The updated version is available in SSC within a few days. And from today
on my own site. It will be removed from my own site when it is in SSC
Jens M. Lauritsen, Consultant,MD. PhD.
Accident Analysis Group, Odense University Hospital,
Odense, Denmark e-mail: [email protected]
Initiator of EpiData, see www.epidata.dk for DataEntry in windows (EpiInfo
V6 principle)
For news on Epidata, sign on at: www.epidata.dk/php/maillist.php
Phone +45 6541 2293 Mobile +45 2173 2293 fax: +45 6613 6581
*
* 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/
