Skip to content
DraconiticResearch
Search Sources API
197,787 studies31 sourcesUpdated 25 min ago
PreprintarXiv licencearXiv

Penalized Cause-Specific Hazard Regression under Competing Risks: An Elastic Net Approach Using SANAD Trial Data

  • Avinash Kumar
Open at sourcePDFDOICite

Abstract

Competing risks analysis is an important framework for studying time-to-event data in the presence of multiple, mutually exclusive events. Traditional methods, such as non-parametric estimators and proportional hazards models, may not adequately capture complex hazard patterns or provide stable estimates in practical settings. In this study, we propose a parametric approach based on the generalized exponential regression model for competing risks data. The model allows flexible hazard shapes and incorporates covariate effects through a regression structure. To improve estimation stability, we integrate elastic net regularization into the likelihood framework. The proposed method is evaluated through simulation studies and applied to a real dataset. The results indicate that the elastic net approach provides more stable and precise estimates compared to standard maximum likelihood estimation. Overall, the proposed framework offers a flexible and practical approach for analyzing competing risks data.

Related

  1. Time-Specific Multinomial Logistic Regression for Competing Risks
    Haoxuan, Zhou, Xiaoqiong +3
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  2. A penalized logistic generalized regression estimator
    Grayson W. White, Kelly S. McConville, Cooper S. Schumacher
    arXivPreprintCC BY 4.0Full textStatisticsarXiv
  3. Fast Computation of Nested Cross-Validation for Penalized Regression
    Shuyang Cao, Alex Stringer
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  4. Linking COPD Prevalence with Income Distribution: A Spatial Heterogeneous Compositional Regression via Geographically Weighted Penalized Approach
    Jingwen Deng, Shujie Ma, Sergio J. Rey +1
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  5. A Unified Coordinate Descent Framework for Penalized Longitudinal Log-Contrast Quantile Regression
    Changwei Du, Linhui Li, Huijuan Ma +1
    arXivPreprintarXiv licenceFull textStatisticsarXiv

Cite

BibTeXRIS
BibTeX
@misc{kumar2026penalized,
  title = {{Penalized Cause-Specific Hazard Regression under Competing Risks: An Elastic Net Approach Using SANAD Trial Data}},
  author = {Avinash Kumar},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.37235},
  eprint = {2610.37235},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.37235},
}
RIS
TY  - UNPB
TI  - Penalized Cause-Specific Hazard Regression under Competing Risks: An Elastic Net Approach Using SANAD Trial Data
AU  - Avinash Kumar
PY  - 2026
DA  - 2026/10/27
PB  - arXiv
DO  - 10.48550/arxiv.2610.37235
UR  - https://arxiv.org/abs/2610.37235
AB  - Competing risks analysis is an important framework for studying time-to-event data in the presence of multiple, mutually exclusive events. Traditional methods, such as non-parametric estimators and proportional hazards models, may not adequately capture complex hazard patterns or provide stable estimates in practical settings. In this study, we propose a parametric approach based on the generalized exponential regression model for competing risks data. The model allows flexible hazard shapes and incorporates covariate effects through a regression structure. To improve estimation stability, we integrate elastic net regularization into the likelihood framework. The proposed method is evaluated through simulation studies and applied to a real dataset. The results indicate that the elastic net approach provides more stable and precise estimates compared to standard maximum likelihood estimation. Overall, the proposed framework offers a flexible and practical approach for analyzing competing risks data.
ER  -