Re: independent right-censor in survival analysis

On 3月21日, 上午9时42分, Richard Ulrich <Rich.Ulr...@xxxxxxxxxxx> wrote:
On Thu, 20 Mar 2008 02:14:01 -0700 (PDT), water

<waterloo2...@xxxxxxxxx> wrote:
How to understand independent right-censor?
Below is a sentence from a book,I can not understand it.
" A right-censoring mechanism is said to be independent if the failure
rates that apply to individuals on trial at each time t>0 are the same
as those that would have applied had there been no censoring."

Thanks in advance!

Assuming this is occurrence or recurrence of disease -
Look for contrast at the opposite side, the *dependency*.
Is the censoring correlated with any aspect/ trait/ symptom
of the disease?

If you are just chopping off follow-up at an arbitrary
number of years, presumably it does not affect their
future outcomes. If you chop off follow-up for one
analysis according to the same calendar date for everyone,
presumably that has nothing to do with their future
outcomes.

However.
If someone is censored because their phone number
no longer works, and you lost them "administratively"....
Could that have been because of illness, loss of job,
etc., which might be the fault of disease? Right-
censoring might indicate traits, etc., that (if you had a
group like this) would result in a much higher rate of
disease ("failure").

Hope that says it well enough.

--
Rich Ulrich

http://www.pitt.edu/~wpilib/index.html

If the censor is independent, we have
P{T in [t,t+h) | x,T >= t } P{T in [t,t+h) | x,T >=
t ,Y(t) = 1}
lim ---------------------------- = lim
------------------------------------
h->0 h h->0
h

here T is failure time variable, Y(t)=1 indicates that the individual
has neithere failed nor been censored prior to time t.

Why?
.

Relevant Pages

  • Re: independent right-censor in survival analysis
    ... Assuming this is occurrence or recurrence of disease - ... Is the censoring correlated with any aspect/ trait/ symptom ... If you are just chopping off follow-up at an arbitrary ... future outcomes. ...
    (sci.stat.edu)
  • Re: Fitting a random sample to a pdf
    ... John D'Errico wrote: ... No. Censoring of data is a different thing from your ... Suppose my widgets have an unknown failure time ...
    (comp.soft-sys.matlab)
  • Re: Fitting a random sample to a pdf
    ... I'd probably use an MLE that has the rounding built ... No. Censoring of data is a different thing from your ... Suppose my widgets have an unknown failure time ...
    (comp.soft-sys.matlab)
Loading