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

Re: st: strange confidence intervals in survival analyses


From   jlinhart@stata.com (Jean Marie Linhart, StataCorp.)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: strange confidence intervals in survival analyses
Date   Tue, 17 Jun 2003 13:06:35 -0500

Paula Brentlinger <brentp2@u.washington.edu> wrote:

>     I am trying to calculate mortality rates using Stata 7.0, but am
> having a problem with confidence intervals.  In my survey data set,
> I have a dichotomous outcome variable representing death, and a
> second variable representing person-time.  I have been using "stset"
> to convert the data into survival-time, and "stptime" to calculate
> mortality rates.  This all seems to work very well except when I add
> p-weights.  The confidence intervals I get for the weighted
> estimates seem, well, improbable: e.g. an estimate of 8.2 is
> reported to have a 95% CI of 3.0, 3.0. This problem is especially
> common in analyses with low numbers of deaths.

If Paula could privately mail me her data and a log of what she did, I
will look into this.

A quick test with some data on hand did not yield strange
confidence intervals.

--Jean Marie
jlinhart@stata.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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index