Skip to content
DraconiticResearch
Search Sources API
226,742 studies31 sourcesUpdated 46 min ago
PreprintarXiv licencearXiv

Debiased Machine Learning for Partially Linear Accelerated Failure Time Models

Abstract

The Cox model remains the default for survival analysis, but the proportional hazards assumption is often violated and hazard ratios can be difficult to interpret. Accelerated failure time (AFT) models provide an intuitive time-scale alternative, yet flexible covariate adjustment while preserving valid inference on a target exposure remains challenging. For the partially linear AFT model under right censoring, a rank-based debiased machine learning (DML) framework has not been developed. The rank-based pairwise moment is not Neyman orthogonal, and standard cross-fitting does not directly apply to U-statistics. We develop the first such framework by combining an orthogonalized rank-based U-statistic, a censoring-corrected influence function, and block-pairwise cross-fitting, yielding valid inference for a finite-dimensional exposure effect under flexible machine-learning nuisance estimation. In the primary simulation study, the proposed estimator attains near-nominal coverage under independent and covariate-dependent censoring at rates up to 60%, whereas the nonorthogonal plug-in estimator remains biased. Using All of Us electronic health record data from adults with diabetes and advanced chronic kidney disease, we apply the proposed method to estimate the associations of serum albumin, estimated glomerular filtration rate, and systolic blood pressure with incident cardiovascular outcomes.

Related

  1. Debiased Machine Learning for Count Data: a Partially Linear Poisson Model Based on Neural Networks
    Ningkun Zhou, Bryan E. Shepherd, Qingyan Xiang
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  2. Automatic debiased machine learning for covariate shifts
    Victor Chernozhukov, Michael K. Newey, Whitney K. Newey +2
    BiometrikaPeer-reviewedLicence not statedFull text27 citedStatisticsOpenAlex
  3. Causal Mediation Analysis for Panel Data with Double/Debiased Machine Learning
    Peikai Wu, Zhiguo Xiao
    arXivPreprintCC BY 4.0Full textStatisticsarXiv
  4. A Debiased Machine Learning Single-Imputation Framework for Item Nonresponse in Surveys
    Mehdi Dagdoug, David Haziza
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  5. Covariate Selection for Doubly Robust Double/debiased Machine Learning Estimators for Causal Inference
    Muwon Kwon, Peter M. Steiner
    arXivPreprintarXiv licenceFull textStatisticsarXiv
BibTeX
@misc{okuno2026debiased,
  title = {{Debiased Machine Learning for Partially Linear Accelerated Failure Time Models}},
  author = {Tomoki Okuno and Sijie Zheng and Brendon Chau and Gang Li and Jin J. Zhou and Hua Zhou},
  howpublished = {arXiv},
  year = {2026},
  month = nov,
  doi = {10.48550/arxiv.2608.07841},
  eprint = {2608.07841},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.07841},
}
RIS
TY  - UNPB
TI  - Debiased Machine Learning for Partially Linear Accelerated Failure Time Models
AU  - Tomoki Okuno
AU  - Sijie Zheng
AU  - Brendon Chau
AU  - Gang Li
AU  - Jin J. Zhou
AU  - Hua Zhou
PY  - 2026
DA  - 2026/11/08
PB  - arXiv
DO  - 10.48550/arxiv.2608.07841
UR  - https://arxiv.org/abs/2608.07841
AB  - The Cox model remains the default for survival analysis, but the proportional hazards assumption is often violated and hazard ratios can be difficult to interpret. Accelerated failure time (AFT) models provide an intuitive time-scale alternative, yet flexible covariate adjustment while preserving valid inference on a target exposure remains challenging. For the partially linear AFT model under right censoring, a rank-based debiased machine learning (DML) framework has not been developed. The rank-based pairwise moment is not Neyman orthogonal, and standard cross-fitting does not directly apply to U-statistics. We develop the first such framework by combining an orthogonalized rank-based U-statistic, a censoring-corrected influence function, and block-pairwise cross-fitting, yielding valid inference for a finite-dimensional exposure effect under flexible machine-learning nuisance estimation. In the primary simulation study, the proposed estimator attains near-nominal coverage under independent and covariate-dependent censoring at rates up to 60%, whereas the nonorthogonal plug-in estimator remains biased. Using All of Us electronic health record data from adults with diabetes and advanced chronic kidney disease, we apply the proposed method to estimate the associations of serum albumin, estimated glomerular filtration rate, and systolic blood pressure with incident cardiovascular outcomes.
ER  -