Re: st: interpreting stset

[Steven Samuels <sjhsamuels@earthlink.net>]

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.

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


> Hello,
> I've stset survival data, and received these messages:
> "      12  multiple records at same instant                      
> PROBABLE ERROR
>             (risktime[_n-1]==risktime)
>         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
> mcculloch@pinestreetfoundation.org
> www.pinestreetfoundation.org
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/