EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

Model Validation for Survival Analysis by Smoothed Predictive Likelihood

Chengyuan Lu iD, Hein Putter iD, Mar Rodríguez Girondo, Jelle J Goeman

Statistics in Medicine · 2025-07 · 44 (15-17)

DOI: 10.1002/sim.70193 ↗

At a glance

Journal article in Statistics in Medicine, published online on 2025-07-18. Written by Chengyuan Lu and 3 co-authors. Institution: Biomedical Data Sciences, Leiden University Medical Center, Leiden, the Netherlands. Major MEDLINE topics: Models, Statistical. Cited 1 times so far (Crossref); 22 references. The full text is freely available (open access, CC BY 4.0). Funded by: NWO, Nederlandse Organisatie voor Wetenschappelijk Onderzoek.

AstrattSubjectsDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Assessing the predictive performance is a crucial aspect in survival modeling, essential for model selection, tuning parameter determination, and evaluating additional predictive ability. The predictive log‐likelihood has been recommended as a suitable evaluation measure, particularly for survival models, which generally return entire survival curves rather than point predictions. However, applying predictive likelihood in semiparametric and nonparametric survival models is problematic since the survival curves are step‐functions, which result in zero predictive likelihood when events occur at previously unobserved time points. The most well‐known existing solution, Verweij's predictive partial likelihood, is limited to Cox models. In this article, we propose a novel approach based on nearest‐neighbor kernel smoothing that is usable in general semi‐ and nonparametric survival models. We show that our new method performs competitively with existing methods in the Cox setting while offering broader applicability, including testing for the presence of a frailty term and determining the optimal level of smoothness in penalized additive hazards models.

Subjects

MeSH terms (MEDLINE)

Models, StatisticalHumansLikelihood FunctionsSurvival AnalysisProportional Hazards ModelsComputer Simulation

Author keywords

Brier scoreadditive hazardscross‐validated partial likelihoodcross‐validation

Authors and institutions

  • Biomedical Data Sciences, Leiden University Medical Center, Leiden, the Netherlands.

ORCID profiles: Chengyuan Lu, Hein Putter

Funding

  • NWO
  • Nederlandse Organisatie voor Wetenschappelijk Onderzoek

Cite

APA

Lu, C., Putter, H., Girondo, M. R., & Goeman, J. J. (2025). Model Validation for Survival Analysis by Smoothed Predictive Likelihood. Statistics in Medicine, 44(15-17). https://doi.org/10.1002/sim.70193

Vancouver

Lu C, Putter H, Girondo MR, Goeman JJ. Model Validation for Survival Analysis by Smoothed Predictive Likelihood. Statistics in Medicine. 2025;44(15-17). doi:10.1002/sim.70193

BibTeX
@article{Lu2025,
  author = {Lu, Chengyuan and Putter, Hein and Girondo, Mar Rodríguez and Goeman, Jelle J},
  title = {Model Validation for Survival Analysis by Smoothed Predictive Likelihood},
  journal = {Statistics in Medicine},
  year = {2025},
  volume = {44},
  number = {15-17},
  doi = {10.1002/sim.70193}
}
Start for free →