Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 26, 2026The American Statistician

How Much Information Remains Under Censoring? A Variance Decomposition for the Kaplan–Meier Estimator

View Full Paper
Ask AI
Bookmark
Share

Authors

SNSzilárd Nemes

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates information retention under censoring in survival data, highlighting practical implications.

Key Points

  • This research aims to quantify how much information about survival data remains after censoring occurs.
  • Developed a variance-decomposition framework to analyze the impact of censoring on survival data.
  • Calculated pointwise measures of information retention including variance-inflation factor and effective sample size.
  • Used simulations to validate the calibration of proposed measures and coverage of the censoring-component interval.
  • Demonstrated that effective sample size accurately reflects information retention, even with low risk-set sizes.
  • Pointwise censoring-component intervals showed near-nominal coverage, indicating reliable inference under censoring conditions.
  • Illustrated through clinical trial examples that observed risk-set size may not necessarily correspond with precision of estimates.

Cite This Study

Szilárd Nemes (2026) studied this question.

synapsesocial.com/papers/6a65a337d3aea3239cd7664bhttps://doi.org/10.1080/00031305.2026.2701342
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Estimating Mean and Variance From Right-censored Data2026
  2. 2Proximal survival analysis to handle dependent right censoring2024 · 5 citations
  3. 3On the effect of the Kaplan-Meier estimator’s assumed tail behavior on goodness-of-fit testing2024 · 1 citations
  4. 4Evaluation of a Non-Parametric Penalized Kaplan–Meier Estimator Under Interval-Censored Survival Data2026 · 1 citations
  5. 5Harvesting information from Kaplan-Meier plots: part 1—detecting censoring2024 · 1 citations