Just add a small amount of time to the early failures. For example, change their failure time to 0.001 instead of 0.
Cox does not care about the absolute time, only about the relative failure time (i.e. order of failures)
Ricardo.
Ricardo Ovaldia, MS
Statistician
Oklahoma City, OK
--- On Fri, 6/6/08, Carlo Lazzaro <[email protected]> wrote:
> From: Carlo Lazzaro <[email protected]>
> Subject: st: R: Initial failure in Cox hazard models
> To: [email protected]
> Cc: "'Shirkey, Zachary C'" <[email protected]>
> Date: Friday, June 6, 2008, 12:54 AM
> Dear Zackary,
>
> I miss to get your following statement: but a large number
> of my cases fail
> in the first time period for which they are observed and
> thus are being
> excluded from the analysis.
>
> Why are they excluded? Are they censored cases?
>
> Kind Regards,
>
> Carlo
> -----Messaggio originale-----
> Da: [email protected]
> [mailto:[email protected]] Per conto di
> Shirkey, Zachary
> C
> Inviato: gioved� 5 giugno 2008 20.23
> A: [email protected]
> Oggetto: st: Initial failure in Cox hazard models
>
> I am using stcox to do survival analysis, but a large
> number of my cases
> fail in the first time period for which they are observed
> and thus are being
> excluded from the analysis. I 'm concerned this is
> biasing my results. How
> can I deal with this issue? Is there a statistical
> technique I can use in
> stcox, or do I need to add a t=0 period prior to the
> beginning of each of my
> cases?
>
> Zachary Shirkey
> Assistant Professor
> Political Science Department
> St. John Fisher College
> 124 Basil Hall
> 3690 East Ave.
> Rochester, NY 14618
> (585) 385-5253
>
>
>
> *
> * 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/
>
>
>
> *
> * 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/
*
* 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/
