pith. machine review for the scientific record. sign in

arxiv: 2604.21020 · v1 · submitted 2026-04-22 · 📊 stat.ME

Recognition: unknown

A Functional-Class Meta-Analytic Framework for Quantifying Surrogate Resilience

Emily Hsiao , Layla Parast
Authors on Pith no claims yet

Pith reviewed 2026-05-09 23:16 UTC · model grok-4.3

classification 📊 stat.ME
keywords surrogate markersurrogate paradoxmeta-analysisresilience probabilityfunctional classclinical trialstreatment effecttransportability
0
0 comments X

The pith

A method estimates the probability that a surrogate marker will produce the wrong conclusion about a treatment's effect on the actual outcome.

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

The paper develops a statistical approach that uses results from multiple past studies measuring both a surrogate marker and the true primary outcome to predict risks in a new study where only the surrogate will be observed. It calculates a resilience probability, defined as the chance that the surrogate paradox occurs: the treatment looks beneficial on the surrogate but is actually harmful on the primary outcome. The approach avoids assuming that the exact relationship between surrogate and outcome will carry over unchanged to the new study. Instead it examines how much that relationship could reasonably differ while still being consistent with the observed data from completed studies. This matters for clinical trial design because it offers a concrete number for deciding whether a surrogate is reliable enough to replace a costly or delayed primary endpoint.

Core claim

By treating the conditional mean function linking treatment, surrogate, and primary outcome as belonging to a class of plausible functions that deviate from those estimated in K completed studies, the method produces an estimate of the resilience probability, which is the probability that the sign of the treatment effect on the surrogate disagrees with the sign on the primary outcome in a new study that measures only the surrogate.

What carries the argument

The resilience probability, obtained by integrating over functional deviations from the conditional mean functions observed in the completed studies without requiring direct transportability of those functions.

If this is right

  • Inference about the new study becomes possible even when only surrogate data are collected there.
  • The estimate accounts for uncertainty in how the surrogate-outcome link may change across studies.
  • Simulation studies and the schizophrenia trial application demonstrate that the method can be computed and yields interpretable probabilities.
  • The same framework can be applied to any collection of completed trials that share the surrogate and primary measures.

Where Pith is reading between the lines

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

  • Low resilience probability would support using the surrogate to accelerate future trials that would otherwise require long follow-up for the primary outcome.
  • The approach could be adapted to settings with multiple surrogates or time-to-event primary outcomes by expanding the functional class accordingly.
  • If the functional class is misspecified, the resilience probability may be over- or under-confident, suggesting a need for sensitivity checks across different classes.

Load-bearing premise

The possible ways the new study's conditional mean function can differ from the observed ones can be adequately described and bounded by a chosen class of functions.

What would settle it

In a new study that measures both surrogate and primary outcome, the observed frequency of the surrogate paradox differs substantially from the resilience probability the method estimated using only the surrogate data and the prior studies.

Figures

Figures reproduced from arXiv: 2604.21020 by Emily Hsiao, Layla Parast.

Figure 1
Figure 1. Figure 1: Functional class synthetic data generation; the red points indicate individual [PITH_FULL_IMAGE:figures/full_fig_p023_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of simulation Settings 1-3 which were designed to have a high probabil [PITH_FULL_IMAGE:figures/full_fig_p024_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Illustration of simulation Settings 4-6 which were designed to have a low probability [PITH_FULL_IMAGE:figures/full_fig_p025_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Relationship between the Brief Psychiatric Rating Scale (BPRS) score (surrogate) [PITH_FULL_IMAGE:figures/full_fig_p027_4.png] view at source ↗
read the original abstract

A surrogate marker is a biomarker or other physical measurement used to replace a primary outcome in clinical trials to evaluate a treatment effect when the primary outcome of interest is costly, invasive, or takes a long time to observe. However, replacing a primary outcome with a surrogate can lead to the "surrogate paradox," in which a treatment appears beneficial based on the surrogate but is actually harmful with respect to the primary outcome. In this paper, we propose a functional class-based method to assess resilience to the surrogate paradox in a meta-analytic setting. Our method leverages data from K completed studies in which the surrogate marker and primary outcome have been measured to make inference on a new study in which only the surrogate is measured. We do not assume direct transportability of the conditional mean function from the completed studies to the new study; instead, we consider deviations of functions from those observed in the completed studies to estimate the "resilience probability" i.e., the probability of the surrogate paradox in the new study. We investigate the performance of our proposed method through a simulation study and apply our method to data from clinical trials in schizophrenia.

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

0 major / 3 minor

Summary. The paper proposes a functional-class meta-analytic framework to quantify resilience to the surrogate paradox. It uses data from K completed studies (where both surrogate marker S and primary outcome Y are observed) to define function classes and estimate the 'resilience probability'—the probability that the surrogate paradox occurs—in a new study where only S is measured. The method avoids direct transportability assumptions on the conditional mean E[Y|S] by instead considering deviations of functions from those observed in the completed studies. Performance is assessed via simulation, and the approach is applied to schizophrenia clinical trial data.

Significance. If the central derivations hold, the framework offers a practical advance for surrogate validation in meta-analytic settings by replacing strong transportability assumptions with a quantifiable resilience probability grounded in observed functional deviations across studies. The simulation study and real-data application to schizophrenia trials provide empirical grounding and demonstrate feasibility, which are explicit strengths of the manuscript.

minor comments (3)
  1. [§2.2] §2.2: the definition of the functional deviation parameters and the resulting resilience probability estimator would benefit from a short numerical illustration (e.g., with K=2 and simple linear functions) to clarify how the class is constructed and how the probability is computed from the observed functions.
  2. [§4] §4 (simulation): the reported coverage and bias results for the resilience probability estimator should include a sensitivity check on the choice of K and on the magnitude of the functional deviations; without this, it is difficult to assess robustness of the method's performance claims.
  3. [§5] §5 (application): the estimated resilience probability for the schizophrenia data is presented without accompanying uncertainty quantification (e.g., bootstrap intervals or posterior credible intervals); adding this would strengthen the interpretability of the real-data result.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive assessment of the manuscript, including the recognition of its practical advance in surrogate validation and the empirical support from simulations and the schizophrenia application. We note the recommendation for minor revision and will address any editorial or presentational suggestions in the revised version.

Circularity Check

0 steps flagged

No significant circularity in the derivation chain

full rationale

The proposed functional-class meta-analytic framework grounds its resilience probability estimate in data from K completed studies by quantifying deviations of conditional mean functions from those observed in the completed studies, without assuming direct transportability to the new study. This construction relies on external empirical inputs rather than self-definitional loops, fitted parameters renamed as predictions, or load-bearing self-citations. The central claim remains independent of its own outputs, with the simulation study and schizophrenia trial application providing separate empirical checks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on using completed study data to define a class of deviating functions for inference on a new study, without specifying exact deviation mechanisms or validation criteria.

free parameters (1)
  • functional deviation parameters
    The method requires specification of how functions deviate from observed conditional means in the completed studies, likely involving parameters to bound or sample the functional class.
axioms (1)
  • domain assumption Data from K completed studies can define a relevant class of possible conditional mean functions applicable to a new study
    Invoked when leveraging prior studies to estimate resilience probability without direct transportability of the conditional mean.

pith-pipeline@v0.9.0 · 5491 in / 1440 out tokens · 67095 ms · 2026-05-09T23:16:41.508825+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

300 extracted references · 8 canonical work pages · 1 internal anchor

  1. [1]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=

    Varying-coefficient models , author=. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=. 1993 , publisher=

    1993

  2. [2]

    Biometrika , volume=

    Maximum likelihood estimation of models for residual covariance in spatial regression , author=. Biometrika , volume=. 1984 , publisher=

    1984

  3. [3]

    Statistics in medicine , volume=

    Flexible evaluation of surrogate markers with Bayesian model averaging , author=. Statistics in medicine , volume=. 2024 , publisher=

    2024

  4. [4]

    Approval of aducanumab for Alzheimer disease—the FDA’s perspective , author=. JAMA. 2021 , publisher=

    2021

  5. [5]

    Nature Reviews Neurology , volume=

    Aducanumab and the FDA—where are we now? , author=. Nature Reviews Neurology , volume=. 2021 , publisher=

    2021

  6. [6]

    , author=

    Revisiting FDA approval of aducanumab. , author=. The New England Journal of Medicine , volume=

  7. [7]

    Nature , volume=

    The antibody aducanumab reduces A plaques in Alzheimer’s disease , author=. Nature , volume=. 2016 , publisher=

    2016

  8. [8]

    Fast Track, Breakthrough Therapy, Accelerated Approval, Priority Review: Accelerated Approval , Year =

    FDA , Howpublished =. Fast Track, Breakthrough Therapy, Accelerated Approval, Priority Review: Accelerated Approval , Year =

  9. [9]

    Biogen to Realign Resources for Alzheimer's Disease Franchise , Year =

    Biogen , Howpublished =. Biogen to Realign Resources for Alzheimer's Disease Franchise , Year =

  10. [10]

    The annals of Statistics , volume=

    Statistical estimation in varying coefficient models , author=. The annals of Statistics , volume=. 1999 , publisher=

    1999

  11. [11]

    10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study

    Knowler, William C and Fowler, Sarah E and Hamman, Richard F and Christophi, Costas A and Hoffman, Heather J and Brenneman, Anne T and Brown-Friday, Janet O and Goldberg, Ronald and Venditti, Elizabeth and Nathan, David M and others , date-added =. 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study. ,...

  12. [12]

    A correspondence between Bayesian estimation on stochastic processes and smoothing by splines , volume =

    Kimeldorf, George S and Wahba, Grace , date-added =. A correspondence between Bayesian estimation on stochastic processes and smoothing by splines , volume =. The Annals of Mathematical Statistics , number =

  13. [13]

    Schapire , date-added =

    Yoav Freund and Robert E. Schapire , date-added =. A Decision-Theoretic Generalization of on-Line Learning and an Application to Boosting , year =

  14. [14]

    A genome-wide association scan of tag SNPs identifies a susceptibility variant for colorectal cancer at 8q24.21 , volume =

    Tomlinson, Ian and Webb, Emily and Carvajal-Carmona, Luis and Broderick, Peter and Kemp, Zoe and Spain, Sarah and Penegar, Steven and Chandler, Ian and Gorman, Maggie and Wood, Wendy and Barclay, Ella and Lubbe, Steven and Martin, Lynn and Sellick, Gabrielle and Jaeger, Emma and Hubner, Richard and Wild, Ruth and Rowan, Andrew and Fielding, Sarah and Howa...

  15. [15]

    A genome-wide association study identifies alleles in FGFR2 associated with risk of sporadic postmenopausal breast cancer , volume =

    Hunter, David J and Kraft, Peter and Jacobs, Kevin B and Cox, David G and Yeager, Meredith and Hankinson, Susan E and Wacholder, Sholom and Wang, Zhaoming and Welch, Robert and Hutchinson, Amy and Wang, Junwen and Yu, Kai and Chatterjee, Nilanjan and Orr, Nick and Willett, Walter C and Colditz, Graham A and Ziegler, Regina G and Berg, Christine D and Buys...

  16. [16]

    A Note on Margin-based Loss Functions in Classification , year =

    Yi Lin , date-added =. A Note on Margin-based Loss Functions in Classification , year =

  17. [17]

    A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis , volume =

    Witten, Daniela M and Tibshirani, Robert and Hastie, Trevor , date-added =. A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis , volume =. Biostatistics , number =

  18. [18]

    Biometrics , volume=

    A measure of the proportion of treatment effect explained by a surrogate marker , author=. Biometrics , volume=. 2002 , publisher=

    2002

  19. [19]

    Biometrics , volume=

    A rank-based approach to evaluate a surrogate marker in a small sample setting , author=. Biometrics , volume=. 2024 , publisher=

    2024

  20. [20]

    A positive deviance approach to understanding key features to improving diabetes care in the medical home , volume =

    Gabbay, Robert A and Friedberg, Mark W and Miller-Day, Michelle and Cronholm, Peter F and Adelman, Alan and Schneider, Eric C , date-added =. A positive deviance approach to understanding key features to improving diabetes care in the medical home , volume =. The Annals of Family Medicine , number =

  21. [21]

    Statistical Methods in Medical Research , volume=

    Proportion of treatment effect explained: An overview of interpretations , author=. Statistical Methods in Medical Research , volume=. 2024 , publisher=

    2024

  22. [22]

    and McKay, James D

    Hung, Rayjean J. and McKay, James D. and Gaborieau, Valerie and Boffetta, Paolo and Hashibe, Mia and Zaridze, David and Mukeria, Anush and Szeszenia-Dabrowska, Neonilia and Lissowska, Jolanta and Rudnai, Peter and Fabianova, Eleonora and Mates, Dana and Bencko, Vladimir and Foretova, Lenka and Janout, Vladimir and Chen, Chu and Goodman, Gary and Field, Jo...

  23. [23]

    A translational engine at the national scale: informatics for integrating biology and the bedside , volume =

    Kohane, Isaac S and Churchill, Susanne E and Murphy, Shawn N , date-added =. A translational engine at the national scale: informatics for integrating biology and the bedside , volume =. Journal of the American Medical Informatics Association , number =

  24. [24]

    and Katzenstein, D.A

    Hammer, S.M. and Katzenstein, D.A. and Hughes, M.D. and Gundacker, H. and Schooley, R.T. and Haubrich, R.H. and Henry, W.K. and Lederman, M.M. and Phair, J.P. and Niu, M. and others , date-added =. A trial comparing nucleoside monotherapy with combination therapy in. New England Journal of

  25. [25]

    and Geller, Frank and Sulem, Patrick and Rafnar, Thorunn and Wiste, Anna and Magnusson, Kristinn P

    Thorgeirsson, Thorgeir E. and Geller, Frank and Sulem, Patrick and Rafnar, Thorunn and Wiste, Anna and Magnusson, Kristinn P. and Manolescu, Andrei and Thorleifsson, Gudmar and Stefansson, Hreinn and Ingason, Andres and Stacey, Simon N. and Bergthorsson, Jon T. and Thorlacius, Steinunn and Gudmundsson, Julius and Jonsson, Thorlakur and Jakobsdottir, Margr...

  26. [26]

    Accounting for the black--white wealth gap: a nonparametric approach , volume =

    Barsky, Robert and Bound, John and Charles, Kerwin Ko' and Lupton, Joseph P , date-added =. Accounting for the black--white wealth gap: a nonparametric approach , volume =. Journal of the American statistical Association , number =

  27. [27]

    Adaptive index models for marker-based risk stratification , volume =

    Tian, Lu and Tibshirani, Robert , date-added =. Adaptive index models for marker-based risk stratification , volume =. Biostatistics , number =

  28. [28]

    Advances and emerging opportunities in diabetes research: a Strategic Planning report of the

    NIDDK , date-added =. Advances and emerging opportunities in diabetes research: a Strategic Planning report of the

  29. [29]

    Albuminuria and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic individuals , volume =

    Gerstein, Hertzel C and Mann, Johannes FE and Yi, Qilong and Zinman, Bernard and Dinneen, Sean F and Hoogwerf, Byron and Hall. Albuminuria and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic individuals , volume =. JAMA: the journal of the American Medical Association , number =

  30. [30]

    American

    ADA , date-added =. American. Diabetes Care , number =

  31. [31]

    An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization THOMAS G

    Dietterich, Thomas G , date-added =. An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization THOMAS G. DIETTERICH tgd@ cs. orst. edu Department of Computer Science, Oregon State University, Corvallis, OR 97331 , volume =. Machine learning , pages =

  32. [32]

    An introduction to support vector machines and other kernel-based learning methods , year =

    Cristianini, Nello and Shawe-Taylor, John , date-added =. An introduction to support vector machines and other kernel-based learning methods , year =

  33. [33]

    An introduction to the augmented inverse propensity weighted estimator , volume =

    Glynn, Adam N and Quinn, Kevin M , date-added =. An introduction to the augmented inverse propensity weighted estimator , volume =. Political analysis , number =

  34. [34]

    Analysis of multiple phenotypes , volume =

    Kent, Jack W , date-added =. Analysis of multiple phenotypes , volume =. Genetic epidemiology , number =

  35. [35]

    Approximate residual balancing: debiased inference of average treatment effects in high dimensions , volume =

    Athey, Susan and Imbens, Guido W and Wager, Stefan , date-added =. Approximate residual balancing: debiased inference of average treatment effects in high dimensions , volume =. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , number =

  36. [36]

    Liao, Katherine P and Kurreeman, Fina and Li, Gang and Duclos, Grant and Murphy, Shawn and Guzman, Raul and Cai, Tianxi and Gupta, Namrata and Gainer, Vivian and Schur, Peter and others , date-added =. Associations of autoantibodies, autoimmune risk alleles, and clinical diagnoses from the electronic medical records in rheumatoid arthritis cases and non--...

  37. [37]

    Asymptotics for kernel estimate of sliced inverse regression , volume =

    Zhu, Li-Xing and Fang, Kai-Tai , date-added =. Asymptotics for kernel estimate of sliced inverse regression , volume =. The Annals of Statistics , number =

  38. [38]

    Arthritis and rheumatism , title =

    Liao, KP and Kurreeman, F and Li, G and Duclos, G and Murphy, S and RG, P and Cai, T and Gupta, N and Gainer, V and Schur, P and others , date-added =. Arthritis and rheumatism , title =

  39. [39]

    Calibrating parametric subject-specific risk estimation , volume =

    Cai, Tianxi and Tian, Lu and Uno, Hajime and Solomon, Scott D and Wei, LJ , date-added =. Calibrating parametric subject-specific risk estimation , volume =. Biometrika , number =

  40. [40]

    Canonical correlation analysis: An overview with application to learning methods , volume =

    Hardoon, David R and Szedmak, Sandor and Shawe-Taylor, John , date-added =. Canonical correlation analysis: An overview with application to learning methods , volume =. Neural Computation , number =

  41. [41]

    Canonical correlation analysis: Uses and interpretation , year =

    Thompson, Bruce , date-added =. Canonical correlation analysis: Uses and interpretation , year =

  42. [42]

    Childhood adversities and adult psychiatric disorders in the national comorbidity survey replication I: associations with first onset of DSM-IV disorders , volume =

    Green, Jennifer Greif and McLaughlin, Katie A and Berglund, Patricia A and Gruber, Michael J and Sampson, Nancy A and Zaslavsky, Alan M and Kessler, Ronald C , date-added =. Childhood adversities and adult psychiatric disorders in the national comorbidity survey replication I: associations with first onset of DSM-IV disorders , volume =. Archives of gener...

  43. [43]

    Childhood adversity and adult psychiatric disorder in the US National Comorbidity Survey , volume =

    Kessler, Ronald C and Davis, Christopher G and Kendler, Kenneth S , date-added =. Childhood adversity and adult psychiatric disorder in the US National Comorbidity Survey , volume =. Psychological medicine , number =

  44. [44]

    Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks , volume =

    Khan, Javed and Wei, Jun S and Ringner, Markus and Saal, Lao H and Ladanyi, Marc and Westermann, Frank and Berthold, Frank and Schwab, Manfred and Antonescu, Cristina R and Peterson, Carsten and others , date-added =. Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks , volume =. Nature medic...

  45. [45]

    Clinical characteristics of pulmonary embolism]

    Riedel, M , date-added =. Clinical characteristics of pulmonary embolism]. , volume =. Vnitr n

  46. [46]

    Laporte, Silvy and Mismetti, Patrick and D. Clinical Predictors for Fatal Pulmonary Embolism in 15 520 Patients With Venous Thromboembolism Findings From the Registro Informatizado de la Enfermedad TromboEmbolica venosa (RIETE) Registry , volume =. Circulation , number =

  47. [47]

    Colorectal Adenomas A Prototype for the Use of Surrogate End Points in the Development of Cancer Prevention Drugs , volume =

    Kelloff, Gary J and Schilsky, Richard L and Alberts, David S and Day, Robert W and Guyton, Kathryn Z and Pearce, Homer L and Peck, Jonathan C and Phillips, Robert and Sigman, Caroline C , date-added =. Colorectal Adenomas A Prototype for the Use of Surrogate End Points in the Development of Cancer Prevention Drugs , volume =. Clinical cancer research , number =

  48. [48]

    Common variants on chromosomes 2q35 and 16q12 confer susceptibility to estrogen receptor-positive breast cancer , volume =

    Stacey, Simon N and Manolescu, Andrei and Sulem, Patrick and Rafnar, Thorunn and Gudmundsson, Julius and Gudjonsson, Sigurjon A and Masson, Gisli and Jakobsdottir, Margret and Thorlacius, Steinunn and Helgason, Agnar and Aben, Katja K and Strobbe, Luc J and Albers-Akkers, Marjo T and Swinkels, Dorine W and Henderson, Brian E and Kolonel, Laurence N and Le...

  49. [49]

    Comparing and combining biomarkers as principle surrogates for time-to-event clinical endpoints , volume =

    Gabriel, Erin E and Sachs, Michael C and Gilbert, Peter B , date-added =. Comparing and combining biomarkers as principle surrogates for time-to-event clinical endpoints , volume =. Statistics in medicine , number =

  50. [50]

    Comparing biomarkers as principal surrogate endpoints , volume =

    Huang, Ying and Gilbert, Peter B , date-added =. Comparing biomarkers as principal surrogate endpoints , volume =. Biometrics , number =

  51. [51]

    and Ferrell, R

    Zerba, K. and Ferrell, R. and Sing, C. , date-added =. Human genetics , number =

  52. [52]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=

    Criteria for surrogate end points based on causal distributions , author=. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=. 2010 , publisher=

    2010

  53. [53]

    arXiv preprint arXiv:2311.01453 , year=

    Ppi++: Efficient prediction-powered inference , author=. arXiv preprint arXiv:2311.01453 , year=

    work page arXiv

  54. [54]

    A unified framework for semiparametrically efficient semi-supervised learning.arXiv preprint arXiv:2502.17741,

    A Unified Framework for Semiparametrically Efficient Semi-Supervised Learning , author=. arXiv preprint arXiv:2502.17741 , year=

    work page arXiv

  55. [55]

    IEEE transactions on pattern analysis and machine intelligence , volume=

    A review of domain adaptation without target labels , author=. IEEE transactions on pattern analysis and machine intelligence , volume=. 2019 , publisher=

    2019

  56. [56]

    BMC Medical Informatics and Decision Making , volume=

    Treatment effect prediction with adversarial deep learning using electronic health records , author=. BMC Medical Informatics and Decision Making , volume=. 2020 , publisher=

    2020

  57. [57]

    International conference on learning representations , year=

    GANITE: Estimation of individualized treatment effects using generative adversarial nets , author=. International conference on learning representations , year=

  58. [58]

    Current HIV research , volume=

    Nucleotide reverse transcriptase inhibitors: a thorough review, present status and future perspective as HIV therapeutics , author=. Current HIV research , volume=. 2017 , publisher=

    2017

  59. [59]

    The Lancet , volume=

    Life expectancy of individuals on combination antiretroviral therapy in high-income countries: a collaborative analysis of 14 cohort studies , author=. The Lancet , volume=. 2008 , publisher=

    2008

  60. [60]

    HIV/AIDS-Research and palliative care , pages=

    HIV protease inhibitors: a review of molecular selectivity and toxicity , author=. HIV/AIDS-Research and palliative care , pages=. 2015 , publisher=

    2015

  61. [61]

    Fahey , author=

    A Brief Chronicle of CD4 as a Biomarker for HIV/AIDS: A Tribute to the Memory of John L. Fahey , author=. Onco Therapeutics , volume=. 2015 , publisher=

    2015

  62. [62]

    Data Mining and Knowledge Discovery , volume=

    Adversarial balancing-based representation learning for causal effect inference with observational data , author=. Data Mining and Knowledge Discovery , volume=. 2021 , publisher=

    2021

  63. [63]

    2009 , publisher=

    Random fields and geometry , author=. 2009 , publisher=

    2009

  64. [64]

    Advances in Neural Information Processing Systems , volume=

    Estimating the effects of continuous-valued interventions using generative adversarial networks , author=. Advances in Neural Information Processing Systems , volume=

  65. [65]

    JCI insight , volume=

    Cell-associated HIV-1 RNA predicts viral rebound and disease progression after discontinuation of temporary early ART , author=. JCI insight , volume=

  66. [66]

    2000 , publisher=

    Asymptotic statistics , author=. 2000 , publisher=

    2000

  67. [67]

    Joint European conference on machine learning and knowledge discovery in databases , pages=

    Toward an understanding of adversarial examples in clinical trials , author=. Joint European conference on machine learning and knowledge discovery in databases , pages=. 2018 , organization=

    2018

  68. [68]

    Proceedings of the twenty-first international conference on Machine learning , pages=

    Learning and evaluating classifiers under sample selection bias , author=. Proceedings of the twenty-first international conference on Machine learning , pages=

  69. [69]

    Theoretical Computer Science , volume=

    Domain adaptation and sample bias correction theory and algorithm for regression , author=. Theoretical Computer Science , volume=. 2014 , publisher=

    2014

  70. [70]

    Advances in data science and information engineering: proceedings from ICDATA 2020 and IKE 2020 , pages=

    A brief review of domain adaptation , author=. Advances in data science and information engineering: proceedings from ICDATA 2020 and IKE 2020 , pages=. 2021 , publisher=

    2020

  71. [71]

    Science , volume=

    Prediction-powered inference , author=. Science , volume=. 2023 , publisher=

    2023

  72. [72]

    Pharmaceutical Statistics: The Journal of Applied Statistics in the Pharmaceutical Industry , volume=

    Surrogate threshold effect: an alternative measure for meta-analytic surrogate endpoint validation , author=. Pharmaceutical Statistics: The Journal of Applied Statistics in the Pharmaceutical Industry , volume=. 2006 , publisher=

    2006

  73. [73]

    Artificial Intelligence and Statistics , pages=

    Robust covariate shift regression , author=. Artificial Intelligence and Statistics , pages=. 2016 , organization=

    2016

  74. [74]

    Advances in neural information processing systems , volume=

    Conformal prediction under covariate shift , author=. Advances in neural information processing systems , volume=

  75. [75]

    , author=

    A tutorial on conformal prediction. , author=. Journal of Machine Learning Research , volume=

  76. [76]

    Journal of statistical planning and inference , volume=

    Improving predictive inference under covariate shift by weighting the log-likelihood function , author=. Journal of statistical planning and inference , volume=. 2000 , publisher=

    2000

  77. [77]

    Considerations in the evaluation of surrogate endpoints in clinical trials: summary of a National Institutes of Health workshop , volume =

    De Gruttola, Victor G and Clax, Pamela and DeMets, David L and Downing, Gregory J and Ellenberg, Susan S and Friedman, Lawrence and Gail, Mitchell H and Prentice, Ross and Wittes, Janet and Zeger, Scott L , date-added =. Considerations in the evaluation of surrogate endpoints in clinical trials: summary of a National Institutes of Health workshop , volume...

  78. [78]

    Constructing biological knowledge bases by extracting information from text sources

    Craven, Mark and Kumlien, Johan and others , booktitle =. Constructing biological knowledge bases by extracting information from text sources. , volume =

  79. [79]

    Contribution to the discussion of the paper by DR Cox , volume =

    Breslow, NE , date-added =. Contribution to the discussion of the paper by DR Cox , volume =. Journal of the Royal Statistical Society, Series B , number =

  80. [80]

    Counterfactual links to the proportion of treatment effect explained by a surrogate marker , volume =

    Taylor, Jeremy MG and Wang, Yue and Thi. Counterfactual links to the proportion of treatment effect explained by a surrogate marker , volume =. Biometrics , number =

Showing first 80 references.