On Fri, 16 May 2003 10:43:09 +0100 "Griffin, J." <[email protected]>
wrote:
> STATA allows an analysis using data in the counting process notation:
> stset stop,failure(dead=1) entry(start) id(id)
> where start and stop are the start and end of each time interval, and
where there can be many (non-overlapping) intervals for a subject. > Is
it valid to use streg with data in this form, or should it only be used
with stcox? I.e. is the likelihood calculated in the correct way when
streg is used? > Jamie Griffin.
Sorry, but I don't understand the nature of your data set from what
you write. Do you have (a) repeated spells per subject, or (b) one
spell per subject, but the survival time data are grouped into
intervals (perhaps of equal length)? And you have organised your data
into -long- format, with one record per interval per subject. If case
(b), you have "interval censoring", and should use appropriate discrete
time regression methods rather than -stcox- or -streg-. They're
straightforward. See the entry under "discrete" in version 8 [ST]
manual.
Stephen
----------------------
Professor Stephen P. Jenkins <[email protected]>
Institute for Social and Economic Research (ISER)
University of Essex, Colchester, CO4 3SQ, UK
Tel: +44 (0)1206 873374. Fax: +44 (0)1206 873151.
http://www.iser.essex.ac.uk
*
* 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/
