[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: interpreting stset

From   Steven Samuels <>
Subject   Re: st: interpreting stset
Date   Wed, 13 Aug 2008 15:35:46 -0400

Michael--the second message is probably informational, if you have delayed (left-truncated) entry. For example in cross-sectional studies of fetal loss, some women may enter a study only in the 2nd or 3rd month of pregnancy, while others enter and fail earlier. The "on or after" of that statement refers to your analysis (follow-up) time, not to the calendar time at which data were collected.

First message: You will have to look to see what Stata meant. "Same instant" will be in the time units in which you record the data. As such, records tied on weeks, say, might not be tied if the units are in days.

Errors could arise if multiple records belonging to one person, each representing the date when a covariate changed, occur at the same (follow-up) time. Check by sorting on id and time and identifying tied times. If in fact there are multiple records at the same time instant, you may have a contradictory data: the value of a covariate is changed on one of the records but not on the others. You will need to resolve these.

On Aug 13, 2008, at 1:58 PMt, Michael McCulloch wrote:

I've stset survival data, and received these messages:
" 12 multiple records at same instant PROBABLE ERROR
22 obs. begin on or after (first) failure?"

I'd like to ask for help in understanding what is meant by "same instant".
And, since subjects were recruited over several years' time, I'm not
sure why it would be a problem if observations begin on or after
(first) failure.


Best wishes,
Michael McCulloch

Pine Street Foundation
124 Pine St., San Anselmo, CA 94960-2674
Tel: (415) 407-1357
Fax: (415) 485-1065

* For searches and help try:
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index