st: Date: Fri, 4 Dec 2009 08:08:50 -0600

["Jabr, Wael M" <wael.jabr@student.utdallas.edu>]

Hello all,

I am trying to use the conditional risk set model (time from entry) for multiple failure time analysis.
I am referring to the Stata document by Mario Cleves and Isabel Cañette (http://www.stata.com/support/faqs/stat/stmfail.html).

While this document says that the input data for analysis can be either event data (each record shows the info upon failure or censoring) or time data (each record tracks the individual on a monthly/yearly basis), the examples shown are all about event data.

In my case, I track individuals on a monthly basis and record among other variables some time varying ones 9sample below). I record the time to the first, second, third... death as in the month it occurs.
I tried using the stsetting as in the document mentioned above and I keep getting the error message:
multiple records at same instant    PROBABLE ERROR

Can someone advise on how to overcome this or provide me with a reference that specifically use time data instead of event data.

Thanks

Month  State   Event   Time0   Time   Risk Sequence
M1      I1      0       1       1       1
M2      I1      0       2       2       1
M3      I1      0       3       3       1
M4      I1      0       4       4       1
M5      I1      0       5       5       1
M6      I1      0       6       6       1
M7      I1      0       7       7       1
M8      I1      0       8       8       1

M1      I2      0       1       1       1
M2      I2      1       2       2       1
M3      I2      1       3       1       2
M4      I2      0       4       1       3
M5      I2      0       5       2       3
M6      I2      1       6       3       3
M7      I2      0       7       1       4
M8      I2      0       8       2       4

*
*   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/