statistics

The Kaplan-Meier estimator as a limit function of Efron's self-consistency iterations

arXiv:2607.27068

summary

The paper proves that the iterative algorithm proposed by Efron indeed converges to the Kaplan‑Meier estimator, showing it is the limit of his self‑consistency iterations.

Abstract

In 1967 Efron showed that the Kaplan-Meier estimator can be defined as a fixed point of a certain mapping, calling this property ``self-consistency''. He also proposed an iterative algorithm for approximating this fixed point and suggested its convergence. The purpose of the present note is to prove this fact.

Topics & keywords