Detecting Early and Late Divergences in Survival Curves Using Nonparametric Effect Measures
Abstract
Clinical trials often show treatment curves that diverge early and converge later, or vice versa patterns that are poorly captured by the proportional-hazards assumption. We develop a joint inferential framework for two nonparametric functionals of censored survival data: the Kaplan--Meier-based Mann--Whitney effect and a novel temporal contrast separating early and late differences. The approach provides interpretable, probability-scale effect measures and enables joint inference for global and temporal contrasts under right censoring. In simulation studies, the method outperforms the log-rank test under non-proportional hazards while maintaining nominal type-I error. A real-world application illustrates how the temporal contrast reveals clinically meaningful early treatment advantages that remain hidden in standard analyses
Related
- SMOOTH DIVERGENCES ESTIMATION FOR CENSORED SURVIVAL DATA
- SCAN: Sequentially Detecting Change-points via Adaptive Nonparametric Inference
- Nonparametric Estimation of Event-Free Survival for Data with Left-Truncated Death and Intermittently Assessed Nonfatal Events
- Detecting adverse high-order drug combinations from individual case safety reports using computational statistics on disproportionality measures
- Spin structures and measures on the moduli space of curves
BibTeX
@misc{langthaler2026detecting,
title = {{Detecting Early and Late Divergences in Survival Curves Using Nonparametric Effect Measures}},
author = {Patrick B. Langthaler and Jun Ma and Jonas Beck},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.02596},
eprint = {2609.02596},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.02596},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Detecting Early and Late Divergences in Survival Curves Using Nonparametric Effect Measures
AU - Patrick B. Langthaler
AU - Jun Ma
AU - Jonas Beck
PY - 2026
DA - 2026/09/02
PB - arXiv
DO - 10.48550/arxiv.2609.02596
UR - https://arxiv.org/abs/2609.02596
AB - Clinical trials often show treatment curves that diverge early and converge later, or vice versa patterns that are poorly captured by the proportional-hazards assumption. We develop a joint inferential framework for two nonparametric functionals of censored survival data: the Kaplan--Meier-based Mann--Whitney effect and a novel temporal contrast separating early and late differences. The approach provides interpretable, probability-scale effect measures and enables joint inference for global and temporal contrasts under right censoring. In simulation studies, the method outperforms the log-rank test under non-proportional hazards while maintaining nominal type-I error. A real-world application illustrates how the temporal contrast reveals clinically meaningful early treatment advantages that remain hidden in standard analyses
ER -