Michael
If you are willing to assume a parametric model, say F(t) = P(T < t)
where T = time to event, you could impose another model for informative
censoring conditional on the (unobserved) value of T: P(censored|T = t)
= g(t) where g(t) and F(t) are known up to a set of parameters and also
could depend on covariates. Then you could attempt to estimate the
parameters by maximimum likehood. You would have to write your own
ml-program to do this. I have done something similar to this for
informative dropout (not a time-to-event model). See
Feiveson, A. H., Metter, E. J., and Paloski, W. H. (2002) "A Statistical
Model for Interpreting Computerized Dynamic Posturography Data", IEEE
Transactions in Biomedical Engineering 49, 300-309.
Al Feiveson
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Maarten buis
Sent: Thursday, February 09, 2006 1:21 PM
To: [email protected]
Subject: Re: st: informative censoring
Michael:
Informative censoring means that the fact that you are censored tells
you something about the hazard rate that you will experience the event
you study. For instance if you study how long elderly people live in a
panel, and respondents choose not to participate in the panel ones they
get gravely ill, than the fact that they are censored tells you
something about the hazard of experience the event (die). If on the
other hand people are censored because you decided to stop collecting
data after five years in the field, than the fact that they are censored
is likely to be unrelated to the probability of experiencing the event.
So deciding that censoring is informative or not is usually done using
information outside the data. It is very hard to draw such conclusions
form the data itself, since in order to know whether the censoring tells
you something about the hazard you need to have data about your
respondents after they were censored, and if you had that than they
wouldn't be censored...
HTH,
Maarten
--- Michael McCulloch <[email protected]> wrote:
> Are there methods in stata for assessing informative censoring, i.e.
> whether it is present, and how to estimate the survival function if
> informative censoring is present.
-----------------------------------------
between 1/2/2006 and 31/3/2006 I will be visiting the UCLA, during this
time the best way to reach me is by email
Maarten L. Buis
Department of Social Research Methodology Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting adress:
Buitenveldertselaan 3 (Metropolitan), room Z214
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
To help you stay safe and secure online, we've developed the all new
Yahoo! Security Centre. http://uk.security.yahoo.com
*
* 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/
