I'm still not exactly clear what you've done, but I think that you have
miss-specified the survival time.
If you look at the example data that you provided, for ID=1 there are
two records the first with duration 12 months and the second with
duration 17 months. Both of these have been calculated by subtracting
firstyearentry from yearexit. The stset command will therefore
interpret this as a total survival of 29 months for ID=1.
This is not the case, the total survival is only 17 months.
The survival duration should be calculated by subtracting yearentry from
yearexit.
By the "dot in stsum", I assume you mean the missing 50th and 75th
percentiles of the survival distribution.
If the overall survival is better than 50%, then there will be no 50th
or 75th percentile, so they will be missing.
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of
[email protected]
Sent: Monday, 17 March 2008 5:49 PM
To: [email protected]
Subject: Re: st: RE: stset and stsum problem
Dear Kieran,
my problem is that I have dot in stsum when use all observations and
streg have strange coefficients
thanks a lot
Quoting Kieran McCaul <[email protected]>:
> What exactly is the problem?
>
> ______________________________________________
> Kieran McCaul MPH PhD
> WA Centre for Health & Ageing (M573)
> University of Western Australia
> Level 6, Ainslie House
> 48 Murray St
> Perth 6000
> Phone: (08) 9224-2140
> Phone: -61-8-9224-2140
> email: [email protected]
> http://myprofile.cos.com/mccaul
> _______________________________________________
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> [email protected]
> Sent: Sunday, 16 March 2008 3:44 AM
> To: [email protected]
> Subject: st: stset and stsum problem
>
> Dear statlist,
> I have tried to set my data for a survival analysis, but I have a
> problem when use stset and put all the observations censored.
> My dataset is this:
> id firstyearentry yearentry yearexit duration canvi
> 1 3/2001 3/2001 3/2002 12 0
> 1 3/2001 4/2002 9/2002 17 1
> 2 8/2002 8/2002 11/2002 3 0
> 2 8/2002 12/2002 5/2003 9 0
> 3 .....
> ...
>
> The command that I use is:
> stset dur, f(canvi==1) id(id)
> If I put all the observation censored and not with stsum I have this:
> stsum;
>
> failure _d: canvi
> analysis time _t: dur
> id: id
>
> | incidence no. of |------ Survival
> time -----|
> | time at risk rate subjects 25% 50%
> 75%
>
---------+--------------------------------------------------------------
> -------
> total | 10661543 .0002694 34301 1025 .
> .
>
>
> If I take only the observation with failure ==1 I have this:
>
> failure _d: canvi
> analysis time _t: dur
> id: id
>
> | incidence no. of |------ Survival
> time -----|
> | time at risk rate subjects 25% 50%
> 75%
>
---------+--------------------------------------------------------------
> -------
> total | 960671 .0030687 2951 131 283
> 469
>
> Someone can help me to understand where is the error?
> Thanks in advance.
> best,
> catia
>
>
>
> *
> * 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/
>
Catia Nicodemo
*
* 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/
