I'm using stci to estimate median survival time from the following
(multiple-record) dataset.
After:
stset `datevar', id(id) f(`var'f=2) sc(`scale') origin(`var'f=1)
id: id
failure event: hpvhf == 2
obs. time interval: (vdate[_n-1], vdate]
exit on or before: failure
t for analysis: (time-origin)/30.4
origin: hpvhf==1
------------------------------------------------------------------------
------
12596 total obs.
9925 ignored because never entered
1582 obs. end on or before enter()
581 obs. begin on or after (first) failure
------------------------------------------------------------------------
------
508 obs. remaining, representing
351 subjects
288 failures in single failure-per-subject data
3644.836 total analysis time at risk, at risk from t = 0
earliest observed entry t = 0
last observed exit t = 62.13816
. stci, median
failure _d: hpvhf == 2
analysis time _t: (vdate-origin)/30.4
origin: hpvhf==1
id: id
| no. of
| subjects 50% Std. Err. [95% Conf.
Interval]
-------------+----------------------------------------------------------
---
total | 351 7.368421 1.110286 6.21711
10.8224
Does anybody know why the CI interval (6.2-10.8) is not balanced around
the median (7.4)? The data came from a cohort study with multiple
follow-up visits (every 4 months).
Many thanks,
Salah Mahmud
*
* 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/
