Re: st: Cox model with a segmented time-dependent covariate

[Marta Garcia-Granero <mgarciagranero@gmail.com>]

Hi Billy:

I never said I had limited success. I've had no success at all. I 
thought that maybe I had to restructure the dataset from wide to long, 
and declare survival settings with multiple rows per patient, and the 
declare that birrubin levels varied with time (s, but then I got an 
error message:

reshape long lbr t, i(patient) j(testday)
replace t=0 if missing(t)
stset time, id(patient) failure(status==1)

                id:  patient
     failure event:  status == 1
obs. time interval:  (time[_n-1], time]
 exit on or before:  failure

------------------------------------------------------------------------------
       36  total obs.
       36  multiple records at same instant PROBABLE ERROR
           (time[_n-1]==time)
------------------------------------------------------------------------------
        0  obs. remaining, representing
        0  subjects
        0  failures in single failure-per-subject data
        0  total analysis time at risk, at risk from t =         0
                             earliest observed entry t =         .
                                  last observed exit t =         .

stvary lbr

no observations
r(2000);


Probably I don't understand those commands at all, since they have a 
very different logic than those in SPSS. I'm not complaining, I like a 
lot the way Stata handles survival analysis, but I have been able to do 
that task.

Regards,
Marta
El 15/01/2014 19:50, William Buchanan escribió:
> Again, it would be very helpful if you provided the syntax (in _Stata_) that you've already used with limited success if you'd like people to provide you with meaningful/useful answers to your query.  Although showing the SPSS equivalent may be helpful for some Stata users, it's unlikely to mean much to the majority of Stata users who may be able to better explain where things went wrong for you if you provide the syntax that you've been trying.
>
> HTH,
> Billy
>
> On Jan 15, 2014, at 12:47 PM, Marta Garcia-Granero <mgarciagranero@gmail.com> wrote:
>
> > El 15/01/2014 19:42, William Buchanan escribió:
> > > Please provide the exact syntax you entered into Stata in addition to the output.
> > The output is SPSS' not Stata's. I have not been able to replicate the results in Stata, that's why I'm asking for help.
> >
> > Regards,
> > Marta

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/