I've been using the R package 'survival' recently.

I understand the way to read the survival curves is given time X what is the percent of widgets still in the field Y. And I can get a confidence interval around Y.

But I've been asked if I can go the other way around. Given a percent still in field Y, can I get a time X (and a confidence interval around X) that matches?

I'm looking for either a function in the R package 'survival', or the theory behind generating this.


This is called prediction of the remaining life. There are many methods to do this, parametric and semiparametric. See for instance:

Prediction of remaining life of power transformers based on left truncated and right censored lifetime data

The idea is to estimate the probability that individual $j$ will fail/die at time $t$ given that it has survived until $t_j$. This is $P(T\leq t\vert T>t_j)$, with $t>t_j$.


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.