0
$\begingroup$

In the context of physical asset management, I have read that the Kaplan-Meier estimate is a useful tool for estimating the survival function, based on failure dates for a set of items which were all installed at the same time.

If I have the failure dates for a set of items which were $\textbf{not}$ necessarily installed at the same time and I have their installation dates, what methods can I use to estimate the survival function?

$\endgroup$
1
  • $\begingroup$ @StatsPlease There is no left-truncation. Before installation, items are not at risk at all..... After installation, items are under observation immediately and there is no delay between risk onset and start of the observation. $\endgroup$ – jujae Feb 15 '17 at 16:51
1
$\begingroup$

There are two options.The Kaplan-Meier survival function is still valid. The key to the estimation is right censoring and not the time the case began. Also parametric families such as the exponential survival function is possible.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.