I am having trouble posting to the list -- I tried to post this yesterday
but I haven't seen it appear... so, I'm trying once more.
I have noticed a slight discrepancy in the results from -stsum- and -stci-
in failure-time data. Suppose that I have the following survival dataset
with a total of 12 observations (6 each in 2 treatment groups), which are
all events:
failure event: fail == 1
obs. time interval: (0, t]
exit on or before: failure
------------------------------------------------------------------------------
12 total obs.
0 exclusions
------------------------------------------------------------------------------
12 obs. remaining, representing
12 failures in single record/single failure data
101 total analysis time at risk, at risk from t = 0
earliest observed entry t = 0
last observed exit t = 16
I am interested in finding out the median survival time for each group.
-stsum- gives 3 and 14, but -stci- gives 3 and 12.
Looking at the data, the median for the second group appears to be 12.
My understanding is that both commands should be using the Kaplan-Meier and
thus should come up with the same answer. Or not? Have I missed something?
. stsum, by(group)
failure _d: fail == 1
analysis time _t: t
| incidence no. of |------ Survival time
-----|
group | time at risk rate subjects 25% 50% 75%
---------+---------------------------------------------------------------------
1
| 21 .2857143 6 2 3 5
2
| 80 .075 6 12 14 15
---------+---------------------------------------------------------------------
total
| 101 .1188119 12 3 6 14
