pith. sign in

arxiv: 2606.18281 · v1 · pith:KCGIGMIBnew · submitted 2026-06-08 · 📊 stat.AP · cs.LG· stat.ML

A Guide to Estimating Conditional Average Treatment Effects in Competing Risks Settings

Daniel Klippert , Sarah Friedrich , Markus Pauly This is my paper

Pith reviewed 2026-06-27 14:30 UTC · model grok-4.3

classification 📊 stat.AP cs.LGstat.ML
keywords conditional average treatment effectscompeting risksmeta-learnerssurvival analysissimulation studyCATE estimationCox regressionrandom survival forests
0
0 comments X

The pith

Comparing six meta-learners provides guidance on estimating conditional average treatment effects from competing risks survival data.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets out to give researchers concrete advice on choosing methods to estimate how a binary treatment affects the risk of one specific event when other competing events are possible. It achieves this by systematically testing six meta-learners that pair either Cox regression or random survival forests for modeling risks with elastic net regression or random forests for modeling the treatment effect directly. A sympathetic reader would care because these estimates support individualized treatment decisions in settings such as medical data where comorbidities can otherwise distort apparent benefits. The comparisons are run across simulation scenarios that vary hazard complexity, treatment heterogeneity, assignment mechanisms, event distributions, and censoring rates. The work also supplies an R package that implements all six approaches for immediate use.

Core claim

The paper claims that meta-learners which combine Cox regression or random survival forests for risk modeling with elastic net regression or random forests for direct CATE modeling, when evaluated across multiple simulation settings differing in hazard complexity, treatment heterogeneity, treatment assignment, event type distribution and censoring, supply practical guidance on model selection for estimating covariate-conditional differences in absolute risk for the event of interest at a fixed time in right-censored competing risks data.

What carries the argument

Meta-learners that adapt standard machine learning algorithms to competing risks by first modeling cause-specific risks and then directly estimating conditional differences in absolute risk.

If this is right

  • Model performance depends on the degree of nonlinearity in hazards and the strength of treatment heterogeneity.
  • Choice among the six meta-learners should be informed by observable data features such as censoring level and balance of event types.
  • Direct modeling of the CATE via random forests can outperform regression-based alternatives when treatment effects are complex.
  • The supplied R package makes the recommended meta-learners immediately available for applied analysis.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The simulation-based ranking could serve as a starting point for method selection in other right-censored settings with multiple event types.
  • Validation on external clinical registries would test whether the simulation-derived guidance transfers to data generated by unknown mechanisms.
  • If the recommended meta-learners prove stable, they could reduce reliance on standard cause-specific hazard models that ignore patient-level treatment effect variation.

Load-bearing premise

The chosen simulation settings differing in hazard complexity, treatment heterogeneity, treatment assignment, event type distribution and censoring are representative enough to yield reliable practical guidance on model selection for real competing risks data.

What would settle it

A real competing risks dataset in which the relative performance ordering of the six meta-learners differs markedly from the patterns seen across the simulated scenarios would invalidate the model-selection recommendations.

read the original abstract

Conditional average treatment effects (CATEs) are central to treatment decision-making in personalized medicine. In competing risks settings, estimating CATEs from survival data allows for patient-specific assessments of treatment effectiveness for a specific event of interest while properly accounting for alternative event types. This distinction is essential in the presence of comorbidities, where competing causes of death may otherwise confound the therapeutic benefit. Focusing on right-censored survival times with binary treatment, we examine CATEs defined as covariate-conditional differences in the absolute risk for the event of interest at a fixed time. To this end, we study meta-learners which adapt machine learning algorithms for CATE estimation in competing risks scenarios. We systematically compare six meta-learners, combining Cox regression or random survival forests for risk modeling with elastic net regression or random forests for direct CATE modeling. To provide practical guidance on model selection, we evaluate their performance in multiple simulation settings, that differ in hazard complexity, treatment heterogeneity, treatment assignment, event type distribution and censoring. To facilitate applied use, we provide the R package, crsurvlearners, which implements all considered approaches.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper develops meta-learners for estimating covariate-conditional differences in absolute risk (CATEs) at a fixed time in right-censored competing-risks data with binary treatment. It compares six combinations (Cox or random survival forests for the risk models crossed with elastic-net or random forests for direct CATE modeling), evaluates them across simulation regimes that vary hazard complexity, treatment heterogeneity, assignment mechanism, event-type distribution and censoring, and supplies the R package crsurvlearners to implement the methods.

Significance. If the simulation results are robust, the work supplies concrete, practitioner-oriented guidance on model choice for CATE estimation under competing risks—an area of direct relevance to personalized-medicine applications with comorbidities. The provision of an open R package that implements all six learners is a clear strength that lowers the barrier to adoption and supports reproducibility.

major comments (1)
  1. [Simulation study (throughout)] The central claim that the comparisons yield reliable practical guidance rests on the assumption that the chosen simulation regimes are representative of real competing-risks data. The manuscript does not report calibration of the designs against empirical features of medical registries (high-dimensional covariates, time-varying effects, dependent censoring), nor sensitivity analyses when those features are altered; this is load-bearing for the guidance conclusions.
minor comments (2)
  1. [Introduction / Methods] Notation for the CATE definition (covariate-conditional difference in absolute risk at fixed time t) should be stated explicitly with the relevant survival and cumulative incidence functions before the meta-learner descriptions.
  2. [Abstract / Methods] The abstract states that six meta-learners are compared, but the precise mapping of each learner to the risk-model / CATE-model pairs is not summarized in a single table; adding such a table would improve readability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive review and for recognizing the practical value of the crsurvlearners package. We respond to the single major comment below.

read point-by-point responses

  1. Referee: [Simulation study (throughout)] The central claim that the comparisons yield reliable practical guidance rests on the assumption that the chosen simulation regimes are representative of real competing-risks data. The manuscript does not report calibration of the designs against empirical features of medical registries (high-dimensional covariates, time-varying effects, dependent censoring), nor sensitivity analyses when those features are altered; this is load-bearing for the guidance conclusions.

    Authors: We agree that the simulation regimes, while spanning a range of hazard complexities, treatment heterogeneities, assignment mechanisms, event-type distributions, and censoring levels, were not calibrated to match summary statistics from any particular medical registry and did not include high-dimensional covariates, time-varying effects, or dependent censoring. This constitutes a genuine limitation on the strength of the practical guidance that can be drawn. In the revised manuscript we will add an explicit limitations subsection that states these omissions and cautions readers against over-generalization. We will also report two new sensitivity experiments: one that increases covariate dimension to p=50 while preserving the existing data-generating processes, and one that introduces dependent censoring via a shared frailty term. These additions will be presented as supplementary material with a brief discussion of how the relative ordering of the six meta-learners changes (or does not change). Time-varying effects will be addressed by clarifying that the random-survival-forest risk models already relax the proportional-hazards assumption, while the Cox-based learners do not; we will not add new time-varying simulations in this revision. revision: partial

Circularity Check

0 steps flagged

No circularity: simulation-based comparison of established meta-learners is self-contained

full rationale

The paper adapts known meta-learners (Cox/RSF risk models crossed with elastic-net/RF CATE models) and evaluates them via simulation under varied regimes. No derivation reduces a claimed result to its own fitted inputs by construction, no self-citation chain is load-bearing for the central comparison, and no uniqueness theorem or ansatz is smuggled in. The simulation design is presented as an empirical benchmark rather than a mathematical identity, so the performance rankings remain falsifiable against external data and do not collapse to the paper's own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Abstract-only review yields no specific free parameters, invented entities, or non-standard axioms; the work rests on standard right-censoring and competing risks assumptions common to survival analysis.

axioms (2)
  • domain assumption Right-censored survival times with binary treatment and competing events
    Explicitly stated as the data setting in the abstract.
  • domain assumption CATE defined via covariate-conditional absolute risk differences at a fixed time
    Definition used to frame the estimation target.

pith-pipeline@v0.9.1-grok · 5733 in / 1276 out tokens · 24325 ms · 2026-06-27T14:30:37.251728+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

87 extracted references · 33 canonical work pages

  1. [1]

    D. M. Kent, E. Steyerberg, and D. Van Klaveren. Personalized evidence based medicine: predictive approaches to heterogeneous treatment effects.BMJ (Clinical research ed.), 363, 2018. doi: 10. 1136/bmj.k4245

    2018

  2. [2]

    E. A. L. Gianicolo, M. Eichler, O. Muensterer, K. Strauch, and M. Blettner. Methods for evaluating causality in observational studies.Deutsches Ärzteblatt international, 117(7):101–107, 2020. doi: 10.3238/arztebl.2020.0101

    work page doi:10.3238/arztebl.2020.0101 2020

  3. [3]

    N. T. Longford and J. A. Nelder. Statistics versus statistical science in the regulatory process. Statistics in Medicine, 18(17-18):2311–2320, 1999

    1999

  4. [4]

    Asiaee, C

    A. Asiaee, C. D. Gravio, C. Beck, Y. Mei, S. Pal, and J. D. Huling. Improving precision of rct-based cate estimation using data borrowing with double calibration.arXiv preprint arXiv:2306.17478, 2025. Comparing Methods for CATE Estimation in Competing Risks47

    arXiv 2025

  5. [5]

    Curth and M

    A. Curth and M. van der Schaar. Understanding the impact of competing events on heterogeneous treatment effect estimation from time-to-event data. InInternational Conference on Artificial In- telligence and Statistics, pages 7961–7980. PMLR, 2023

    2023

  6. [6]

    J. G. Young, M. J. Stensrud, E. J. Tchetgen Tchetgen, and M. A. Hernán. A causal framework for classical statistical estimands in failure-time settings with competing events.Statistics in Medicine, 39(8):1199–1236, 2020

    2020

  7. [7]

    N. A. Schuster, E. O. Hoogendijk, A. A. Kok, J. W. Twisk, and M. W. Heymans. Ignoring competing events in the analysis of survival data may lead to biased results: A nonmathematical illustration of competing risk analysis.Journal of Clinical Epidemiology, 122:42–48, 2020. doi: https://doi.org/ 10.1016/j.jclinepi.2020.03.004

    work page doi:10.1016/j.jclinepi.2020.03.004 2020

  8. [8]

    H. Shen, C. Zhang, Y. Song, Z. Huang, Y. Wang, Y. Hou, and Z. Chen. Assessing treatment effects with adjusted restricted mean time lost in observational competing risks data.BMC Medical Research Methodology, 24(1):186, 2024

    2024

  9. [9]

    E. O. Hoogendijk, D. J. H. Deeg, J. Poppelaars, M. van der Horst, M. I. Broese van Groenou, H. C. Comijs, H. R. W. Pasman, N. M. van Schoor, B. Suanet, F. Thomése, T. G. van Tilburg, M. Visser, andM.Huisman. ThelongitudinalagingstudyAmsterdam: Cohortupdate2016andmajorfindings. European Journal of Epidemiology, 31(9):927–945, 2016. doi: 10.1007/s10654-016-0192-3

    work page doi:10.1007/s10654-016-0192-3 2016

  10. [10]

    E. O. Hoogendijk, D. J. H. Deeg, S. de Breij, S. S. Klokgieters, A. A. L. Kok, N. Stringa, E. J. Timmermans, N. M. van Schoor, E. M. van Zutphen, M. van der Horst, J. Poppelaars, P. Malhoe, and M. Huisman. The longitudinal aging study Amsterdam: Cohort update 2019 and additional data collections.European Journal of Epidemiology, 35(1):61–74, 2019

    2019

  11. [11]

    van Walraven and F

    C. van Walraven and F. A. McAlister. Competing risk bias was common in Kaplan–Meier risk estimates published in prominent medical journals.Journal of Clinical Epidemiology, 69:170–173,

  12. [12]

    doi: https://doi.org/10.1016/j.jclinepi.2015.07.006

    ISSN 0895-4356. doi: https://doi.org/10.1016/j.jclinepi.2015.07.006

    work page doi:10.1016/j.jclinepi.2015.07.006 2015

  13. [13]

    Schumacher, K

    M. Schumacher, K. Ohneberg, and J. Beyersmann. Competing risk bias was common in a prominent medical journal.Journal of Clinical Epidemiology, 80:135–136, 2016

    2016

  14. [14]

    Salditt, T

    M. Salditt, T. Eckes, and S. Nestler. A tutorial introduction to heterogeneous treatment effect estimation with meta-learners.Administration and Policy in Mental Health and Mental Health Services Research, pages 1–24, 2023. doi: 10.1007/s10488-023-01303-9

    work page doi:10.1007/s10488-023-01303-9 2023

  15. [15]

    S. R. Künzel, J. S. Sekhon, P. J. Bickel, and B. Yu. Metalearners for estimating heterogeneous treatment effects using machine learning.Proceedings of the National Academy of Sciences, 116 (10):4156–4165, 2019. doi: 10.1073/pnas.1804597116. Comparing Methods for CATE Estimation in Competing Risks48

    work page doi:10.1073/pnas.1804597116 2019

  16. [16]

    M. C. Knaus, M. Lechner, and A. Strittmatter. Machine learning estimation of heterogeneous causal effects: Empirical Monte Carlo evidence.The Econometrics Journal, 24(1):134–161, 06 2020. ISSN 1368-4221. doi: 10.1093/ectj/utaa014

    work page doi:10.1093/ectj/utaa014 2020

  17. [17]

    Caron, G

    A. Caron, G. Baio, and I. Manolopoulou. Estimating individual treatment effects using non- parametric regression models: a review.Journal of the Royal Statistical Society Series A: Statistics in Society, 185(3):1115–1149, 03 2022. ISSN 0964-1998. doi: 10.1111/rssa.12824

    work page doi:10.1111/rssa.12824 2022

  18. [18]

    Y. Xu, N. Ignatiadis, E. Sverdrup, S. Fleming, S. Wager, and N. Shah.Treatment heterogeneity with survival outcomes, pages 445–482. Chapman & Hall/CRC, 2023. doi: 10.1201/9781003102670-21

    work page doi:10.1201/9781003102670-21 2023

  19. [19]

    E. H. Kennedy. Towards optimal doubly robust estimation of heterogeneous causal effects.Electronic Journal of Statistics, 17(2):3008–3049, 2023. doi: 10.1214/23-EJS2157

    work page doi:10.1214/23-ejs2157 2023

  20. [20]

    Powers, J

    S. Powers, J. Qian, K. Jung, A. Schuler, N. H. Shah, T. Hastie, and R. Tibshirani. Some methods for heterogeneous treatment effect estimation in high dimensions.Statistics in Medicine, 37(11): 1767–1787, 2018

    2018

  21. [21]

    Nie and S

    X. Nie and S. Wager. Quasi-oracle estimation of heterogeneous treatment effects.Biometrika, 108 (2):299–319, 2020. doi: 10.1093/biomet/asaa076

    work page doi:10.1093/biomet/asaa076 2020

  22. [22]

    D. R. Cox. Regression models and life-tables.Journal of the Royal Statistical Society. Series B (Methodological), 34(2):187–220, 1972

    1972

  23. [23]

    Tibshirani

    R. Tibshirani. Regression shrinkage and selection via the lasso.Journal of the Royal Statistical Society. Series B (Methodological), 58(1):267–288, 1996. ISSN 00359246

    1996

  24. [24]

    Ishwaran, T

    H. Ishwaran, T. A. Gerds, U. B. Kogalur, R. D. Moore, S. J. Gange, and B. M. Lau. Random survival forests for competing risks.Biostatistics, 15(4):757–773, 2014. doi: 10.1093/biostatistics/kxu010

    work page doi:10.1093/biostatistics/kxu010 2014

  25. [25]

    E., Ercolano, B., & Bonnell, I

    H. Zou and T. Hastie. Regularization and Variable Selection Via the Elastic Net.Journal of the Royal Statistical Society Series B: Statistical Methodology, 67(2):301–320, 2005. doi: 10.1111/j. 1467-9868.2005.00503.x

    work page doi:10.1111/j 2005

  26. [26]

    L. Breiman. Random forests.Machine Learning, 45:5–32, 2001. doi: 10.1023/A:1010933404324

    work page doi:10.1023/a:1010933404324 2001

  27. [27]

    Ishwaran and U

    H. Ishwaran and U. Kogalur.Fast Unified Random Forests for Survival, Regression, and Classifica- tion (RF-SRC), 2023. R package version 3.2.3

    2023

  28. [28]

    B. R. Logan, R. Sparapani, R. E. McCulloch, and P. W. Laud. Decision making and uncertainty quantification for individualized treatments using bayesian additive regression trees.Statistical Methods in Medical Research, 28(4):1079–1093, 2019. doi: 10.1177/0962280217746191. Comparing Methods for CATE Estimation in Competing Risks49

    work page doi:10.1177/0962280217746191 2019

  29. [29]

    L. Hu, J. Ji, and F. Li. Estimating heterogeneous survival treatment effect in observational data using machine learning.Statistics in Medicine, 40(21):4691–4713, 2021

    2021

  30. [30]

    Hu, J.-Y

    L. Hu, J.-Y. Lin, K. Sigel, and M. Kale. Estimating heterogeneous survival treatment effects of lung cancer screening approaches: A causal machine learning analysis.Annals of Epidemiology, 62: 36–42, 2021. doi: 10.1016/j.annepidem.2021.06.008

    work page doi:10.1016/j.annepidem.2021.06.008 2021

  31. [31]

    Klippert

    D. Klippert. An analysis of causal inference methods for competing risks. Unpublished master’s thesis, TU Dortmund University, Dortmund, Germany, June 2024

    2024

  32. [32]

    Athey and G

    S. Athey and G. Imbens. Recursive partitioning for heterogeneous causal effects.Proceedings of the National Academy of Sciences, 113(27):7353–7360, 2016

    2016

  33. [33]

    Wager and S

    S. Wager and S. Athey. Estimation and inference of heterogeneous treatment effects using random forests.Journal of the American Statistical Association, 113(523):1228–1242, 2018

    2018

  34. [34]

    H. A. Chipman, E. I. George, and R. E. McCulloch. Bart: Bayesian additive regression trees.The Annals of Applied Statistics, 4(1):266–298, March 2010

    2010

  35. [35]

    P. R. Hahn, J. S. Murray, and C. M. Carvalho. Bayesian regression tree models for causal inference: Regularization, confounding, and heterogeneous effects (with discussion).Bayesian Analysis, 15(3): 965–1056, 2020

    2020

  36. [36]

    Y. Cui, M. R. Kosorok, E. Sverdrup, S. Wager, and R. Zhu. Estimating heterogeneous treatment effects with right-censored data via causal survival forests.Journal of the Royal Statistical Society Series B: Statistical Methodology, 85(2):179–211, 2023

    2023

  37. [37]

    R. Li, H. Wang, and W. Tu. Robust estimation of heterogeneous treatment effects using electronic health record data.Statistics in Medicine, 40(11):2713–2752, 2021

    2021

  38. [38]

    R. Li, H. Wang, Y. Zhao, J. Su, and W. Tu. Robust estimation of heterogeneous treatment effects: an algorithm-based approach.Communications in Statistics-Simulation and Computation, 52(10): 4981–4998, 2023

    2023

  39. [39]

    Wendling, K

    T. Wendling, K. Jung, A. Callahan, A. Schuler, N. H. Shah, and B. Gallego. Comparing methods for estimation of heterogeneous treatment effects using observational data from health care databases. Statistics in Medicine, 37(23):3309–3324, 2018

    2018

  40. [40]

    Carvalho, A

    C. Carvalho, A. Feller, J. Murray, S. Woody, and D. Yeager. Assessing treatment effect variation in observational studies: Results from a data challenge.Observational Studies, 5(2):21–35, 2019. Comparing Methods for CATE Estimation in Competing Risks50

    2019

  41. [41]

    Curth and M

    A. Curth and M. Van der Schaar. Nonparametric estimation of heterogeneous treatment effects: From theory to learning algorithms. InInternational Conference on Artificial Intelligence and Statistics, pages 1810–1818. PMLR, 2021

    2021

  42. [42]

    D. Jacob. CATE meets ML.Digital Finance, 3:99–148, 2021. doi: 10.1007/s42521-021-00033-7

    work page doi:10.1007/s42521-021-00033-7 2021

  43. [43]

    Zhang, J

    W. Zhang, J. Li, and L. Liu. A unified survey of treatment effect heterogeneity modelling and uplift modelling.ACM Computing Surveys (CSUR), 54(8):1–36, 2021

    2021

  44. [44]

    Bo, J.-H

    N. Bo, J.-H. Jeong, E. Forno, and Y. Ding. Evaluating meta-learners to analyze treatment het- erogeneity in survival data: Application to electronic health records of pediatric asthma care in COVID-19 pandemic.Statistics in Medicine, 44(3-4):e10333, 2025

    2025

  45. [45]

    Möller, E.-M

    M. Möller, E.-M. Wild, W. Tan, and J. Schreyögg. Estimating heterogeneous treatment effects with real-world health data–a scoping review of machine learning methods.Value in Health, 2026

    2026

  46. [46]

    B. M. H. Ozenne, T. H. Scheike, L. Stærk, and T. A. Gerds. On the estimation of average treatment effects with right-censored time to event outcome and competing risks.Biometrical Journal, 62(3): 751–763, 2020

    2020

  47. [47]

    Cantagallo, M

    E. Cantagallo, M. De Backer, M. Kicinski, B. Ozenne, L. Collette, C. Legrand, M. Buyse, and J. Péron. A new measure of treatment effect in clinical trials involving competing risks based on generalized pairwise comparisons.Biometrical Journal, 63(2):272–288, 2021

    2021

  48. [48]

    R. Li, V. B. Talisa, and C.-C. H. Chang. Doubly robust targeted estimation of conditional average treatment effects for time-to-event outcomes with competing risks.arXiv preprint arXiv:2407.18389, 2024

    arXiv 2024

  49. [49]

    Frauen, M

    D. Frauen, M. Schröder, K. Hess, and S. Feuerriegel. Orthogonal survival learners for estimating heterogeneous treatment effects from time-to-event data. InNeurIPS, 2025

    2025

  50. [50]

    Sverdrup and Y

    E. Sverdrup and Y. Cui. Proximal causal learning of conditional average treatment effects. In International Conference on Machine Learning, pages 33285–33298. PMLR, 2023

    2023

  51. [51]

    Kallus, X

    N. Kallus, X. Mao, and A. Zhou. Interval estimation of individual-level causal effects under unob- served confounding. InThe 22nd International Conference on Artificial Intelligence and Statistics, pages 2281–2290. PMLR, 2019

    2019

  52. [52]

    Syrgkanis, V

    V. Syrgkanis, V. Lei, M. Oprescu, M. Hei, K. Battocchi, and G. Lewis. Machine learning estimation of heterogeneous treatment effects with instruments.Advances in Neural Information Processing Systems, 32, 2019. Comparing Methods for CATE Estimation in Competing Risks51

    2019

  53. [53]

    Frauen and S

    D. Frauen and S. Feuerriegel. Estimating individual treatment effects under unobserved confounding using binary instruments.arXiv preprint arXiv:2208.08544, 2023

    arXiv 2023

  54. [54]

    T. Meir, U. Shalit, and M. Gorfine. Heterogeneous treatment effect in time-to-event outcomes: Harnessing censored data with recursively imputed trees. In A. Singh, M. Fazel, D. Hsu, S. Lacoste- Julien, F. Berkenkamp, T. Maharaj, K. Wagstaff, and J. Zhu, editors,Proceedings of the 42nd International Conference on Machine Learning, volume 267 ofProceedings ...

    2025

  55. [55]

    Beyersmann and T

    J. Beyersmann and T. H. Scheike.Classical Regression Models for Competing Risks, pages 157–177. Chapman & Hall/CRC, 2013. doi: 10.1201/b16248

    work page doi:10.1201/b16248 2013

  56. [56]

    J. Neyman. On the application of probability theory to agricultural experiments. Essay on principles. Section 9.Statistical Science, 5(4):465–472, 1990. doi: 10.1214/ss/1177012031

    work page doi:10.1214/ss/1177012031 1990

  57. [57]

    D. B. Rubin. Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology, 66(5):688–701, 1974. doi: 10.1037/h0037350

    work page doi:10.1037/h0037350 1974

  58. [58]

    J. Hill, G. Perrett, and V. Dorie.Machine Learning for Causal Inference, pages 415–444. Chapman & Hall/CRC, 2023. doi: 10.1201/9781003102670-20

    work page doi:10.1201/9781003102670-20 2023

  59. [59]

    B. Neal. Introduction to causal inference.Course Lecture Notes, 2020

    2020

  60. [60]

    R. J. Gray. A class of K-sample tests for comparing the cumulative incidence of a competing risk. The Annals of Statistics, 16(3):1141–1154, 1988. doi: 10.1214/aos/1176350951

    work page doi:10.1214/aos/1176350951 1988

  61. [61]

    O. O. Aalen and S. Johansen. An empirical transition matrix for non-homogeneous Markov chains based on censored observations.Scand. J. Statist., 5(3):141–150, 1978

    1978

  62. [62]

    Y. Xu, N. Ignatiadis, E. Sverdrup, S. Fleming, S. Wager, and N. Shah. survlearners: Metalearners for survival data. 2022. R package version 0.0.1

    2022

  63. [63]

    A. E. Hoerl and R. W. Kennard. Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 12(1):55–67, 1970. doi: 10.1080/00401706.1970.10488634

    work page doi:10.1080/00401706.1970.10488634 1970

  64. [64]

    D. M. Vock, J. Wolfson, S. Bandyopadhyay, G. Adomavicius, P. E. Johnson, G. Vazquez-Benitez, and P. J. O’Connor. Adapting machine learning techniques to censored time-to-event health record data: A general-purpose approach using inverse probability of censoring weighting.Journal of Biomedical Informatics, 61:119–131, 2016. ISSN 1532-0464. doi: 10.1016/j.j...

    work page doi:10.1016/j.jbi.2016.03.009 2016

  65. [65]

    Kogalur, Eugene H

    H. Ishwaran, U. B. Kogalur, E. H. Blackstone, and M. S. Lauer. Random survival forests.The Annals of Applied Statistics, 2(3), 2008. ISSN 1932-6157. doi: 10.1214/08-aoas169. Comparing Methods for CATE Estimation in Competing Risks52

    work page doi:10.1214/08-aoas169 2008

  66. [66]

    E. L. Kaplan and P. Meier. Nonparametric estimation from incomplete observations.Journal of the American statistical association, 53(282):457–481, 1958

    1958

  67. [67]

    J. L. Hill. Bayesian nonparametric modeling for causal inference.Journal of Computational and Graphical Statistics, 20(1):217–240, 2011. doi: 10.1198/jcgs.2010.08162

    work page doi:10.1198/jcgs.2010.08162 2011

  68. [68]

    J. C. Foster, J. M. Taylor, and S. J. Ruberg. Subgroup identification from randomized clinical trial data.Statistics in Medicine, 30(24):2867–2880, 2011

    2011

  69. [69]

    D. G. Horvitz and D. J. Thompson. A generalization of sampling without replacement from a finite universe.Journal of the American statistical Association, 47(260):663–685, 1952

    1952

  70. [70]

    Hirano, G

    K. Hirano, G. W. Imbens, and G. Ridder. Efficient estimation of average treatment effects using the estimated propensity score.Econometrica, 71(4):1161–1189, 2003

    2003

  71. [71]

    P. M. Robinson. Root-n-consistent semiparametric regression.Econometrica, 56(4):931–954, 1988. doi: 10.2307/1912705

    work page doi:10.2307/1912705 1988

  72. [72]

    The Econometrics Journal , volume =

    V. Chernozhukov, D. Chetverikov, M. Demirer, E. Duflo, C. Hansen, W. Newey, and J. Robins. Double/debiased machine learning for treatment and structural parameters.The Econometrics Journal, 21(1):C1–C68, 01 2018. doi: 10.1111/ectj.12097

    work page doi:10.1111/ectj.12097 2018

  73. [73]

    Bender, T

    R. Bender, T. Augustin, and M. Blettner. Generating survival times to simulate Cox proportional hazards models.Statistics in Medicine, 24(11):1713–1723, 2005. doi: 10.1002/sim.2059

    work page doi:10.1002/sim.2059 2005

  74. [74]

    Beyersmann, A

    J. Beyersmann, A. Latouche, A. Buchholz, and M. Schumacher. Simulating competing risks data in survival analysis.Statistics in Medicine, 28(6):956–971, 2009. doi: 10.1002/sim.3516

    work page doi:10.1002/sim.3516 2009

  75. [75]

    M. G. Kendall. The treatment of ties in ranking problems.Biometrika, 33(3):239–251, 11 1945. ISSN 0006-3444. doi: 10.1093/biomet/33.3.239

    work page doi:10.1093/biomet/33.3.239 1945

  76. [76]

    M. G. Kendall. A new measure of rank correlation.Biometrika, 30(1-2):81–93, 1938

    1938

  77. [77]

    Pintilie.Competing risks: A practical perspective

    M. Pintilie.Competing risks: A practical perspective. John Wiley & Sons, 2006

    2006

  78. [78]

    Hodgkin Lymphoma Treatment (PDQ)–Patient Version, 2023

    National Cancer Institute (U.S.). Hodgkin Lymphoma Treatment (PDQ)–Patient Version, 2023. URL https://www.cancer.gov/types/lymphoma/patient/adult-hodgkin-treatment-pdq. Version of November 17, 2023

    2023

  79. [79]

    Focus on extremely late relapses in Hodgkin lymphoma: A single center analysis.Cancers, 18 (5):777, 2026

    A.Liaskas, F.Panitsas, M.K.Angelopoulou, M.Siakantaris, G.A.Pangalis, andT.P.Vassilakopou- los. Focus on extremely late relapses in Hodgkin lymphoma: A single center analysis.Cancers, 18 (5):777, 2026. Comparing Methods for CATE Estimation in Competing Risks53

    2026

  80. [80]

    Rühl and S

    J. Rühl and S. Friedrich. Resampling-based confidence intervals and bands for the average treatment effect in observational studies with competing risks.Statistics and Computing, 34(3), 2024. doi: 10.1007/s11222-024-10420-w

    work page doi:10.1007/s11222-024-10420-w 2024

Showing first 80 references.