pith. sign in

arxiv: 2605.18724 · v1 · pith:V3ZBYLEKnew · submitted 2026-05-18 · 📊 stat.ME

Sensitivity analysis for causal mediation: bridge score, sharp sensitivity bounds, and calibration

Yuki Ohnishi , Fan Li This is my paper

Pith reviewed 2026-05-20 08:03 UTC · model grok-4.3

classification 📊 stat.ME
keywords causal mediationsensitivity analysisbridge scorebalancing scoresequential ignorabilitymediator-outcome confoundingcalibration
0
0 comments X

The pith

The bridge score, built from treatment-specific mediator densities, acts as a balancing score that produces sharp pointwise bounds on the mediator-outcome confounding function using two latent parameters.

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

Causal mediation analysis decomposes total effects into direct and indirect parts but relies on the untestable mediator-stage sequential ignorability assumption. The paper introduces the bridge score as a low-dimensional vector from the two treatment-specific mediator densities at the same mediator value and proves it balances that stage. Conditioning on the bridge score yields a sharp envelope for the unidentified confounding function expressed through two scalar latent parameters. Calibration methods against observed covariates or residual outcome variation turn the bounds into practical sensitivity tools. A scalar reduction plus Bayesian g-computation then delivers posterior inference on the mediation effects while propagating all uncertainty.

Core claim

The bridge score is a balancing score for the mediator stage of sequential ignorability. Conditional on the bridge score, a sharp pointwise envelope exists for the unidentified mediator-outcome confounding function in terms of two interpretable latent confounding parameters. Benchmark calibration against an observed covariate (including a rank-based version) and residual budget calibration make the bounds operational, and a scalar functional reduction combined with Bayesian g-computation propagates uncertainty into posterior draws of the mediation effect estimates.

What carries the argument

The bridge score, a low-dimensional vector formed from the two treatment-specific mediator densities evaluated at a common mediator value, which balances the mediator distributions and permits bounding the unidentified confounding function up to two scalars.

If this is right

  • The pointwise bound on the confounding function becomes usable for sensitivity analysis once the two latent parameters are calibrated to observed covariates or residual outcome variation.
  • A scalar functional reduction converts the pointwise envelope into a form suitable for estimating natural direct and indirect effects.
  • Bayesian g-computation propagates uncertainty from the sensitivity bounds, the calibration, and the observed data into posterior distributions of the mediation effects.
  • The approach applies whenever mediator-outcome confounding is the primary violation of sequential ignorability and the mediator densities can be estimated.

Where Pith is reading between the lines

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

  • The same density-based balancing idea might reduce dimensionality in sensitivity analysis for time-varying mediators if analogous bridge scores can be constructed sequentially.
  • When mediator densities are estimated nonparametrically, the bridge score could serve as a diagnostic for how much observed covariate adjustment already controls mediator-stage confounding.
  • The two-parameter envelope might be compared against existing single-parameter sensitivity methods to quantify the gain from retaining the full pointwise form rather than collapsing early.

Load-bearing premise

The two treatment-specific mediator densities at a fixed mediator value are sufficient to construct a balancing score that renders the unidentified mediator-outcome confounding function identifiable up to two scalar latent parameters.

What would settle it

Simulate data from a known data-generating process with specified mediator-outcome confounding strength, compute the bridge score, set the two latent parameters to their true values, and verify whether the derived envelope always contains the true confounding function value at every mediator point.

Figures

Figures reproduced from arXiv: 2605.18724 by Fan Li, Yuki Ohnishi.

Figure 1
Figure 1. Figure 1: Raw-scale and rank-based benchmark calibration for the immigration-framing [PITH_FULL_IMAGE:figures/full_fig_p017_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Sensitivity parameter sweeps for the immigration framing illustration. Left: [PITH_FULL_IMAGE:figures/full_fig_p018_2.png] view at source ↗
read the original abstract

Causal mediation analysis decomposes the total treatment effect into a portion operating through a hypothesized mediator and a residual direct portion. Identification of natural direct and indirect effects typically rests on the mediator stage of sequential ignorability, which cannot be empirically verified and requires explicit sensitivity analysis. We introduce the \emph{bridge score}, a low-dimensional vector formed from the two treatment-specific mediator densities at a common mediator value, and show that it is a balancing score for the mediator stage of sequential ignorability. Conditional on the bridge score, we then derive a sharp pointwise envelope on the unidentified mediator-outcome confounding function in terms of two interpretable latent confounding parameters. To make the bound operational for sensitivity analysis, we further introduce two calibration approaches. The first is benchmark calibration against an observed covariate, including a rank-based version that is invariant to monotone re-expressions of the benchmark; the second is residual budget calibration based on residual outcome variation. Finally, we show how the pointwise bound can be operationalized for inference through a scalar functional reduction and a Bayesian g-computation algorithm that propagates all sources of uncertainty into posterior draws of the mediation effect estimates.

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 introduces the bridge score, formed as a low-dimensional vector from the two treatment-specific mediator densities evaluated at a common mediator value. It establishes that this score is a balancing score for the mediator stage of sequential ignorability. Conditional on the bridge score, the authors derive a sharp pointwise envelope for the unidentified mediator-outcome confounding function expressed via two scalar latent confounding parameters. Two calibration strategies are proposed: benchmark calibration (including a rank-based variant invariant to monotone transformations) against an observed covariate, and residual budget calibration using residual outcome variation. Inference is operationalized via a scalar functional reduction and a Bayesian g-computation procedure that propagates uncertainty into posterior draws of natural direct and indirect effects.

Significance. If the balancing property and sharpness of the envelope hold, the framework supplies a dimension-reduced, interpretable sensitivity analysis for mediation that is more structured than fully nonparametric approaches while remaining operational through calibration to observables. The explicit construction of the bridge score from observable densities and the provision of both benchmark and residual-budget calibration routes are practical strengths; the Bayesian g-computation step ensures all sources of uncertainty are accounted for in the final mediation-effect posteriors.

major comments (1)
  1. The central claim that the bridge score is a balancing score for the mediator stage of sequential ignorability (and thereby permits a sharp pointwise envelope on the confounding function) rests on the sufficiency of the pair of treatment-specific mediator densities at a fixed mediator value. A concrete verification of this dimension-reduction step—e.g., showing that the conditional independence statements implied by sequential ignorability are preserved after conditioning on the bridge score—would strengthen the load-bearing derivation.
minor comments (2)
  1. The abstract states the central derivation and calibration steps but supplies no equations; adding a brief display of the bridge-score definition and the form of the envelope (even if only in the introduction) would improve readability for readers who begin with the abstract.
  2. Clarify whether the rank-based benchmark calibration remains valid when the benchmark covariate is itself a function of the mediator or treatment; a short remark or counter-example would prevent misapplication.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their positive evaluation and constructive suggestion. We have revised the manuscript to provide a more explicit verification of the bridge score's balancing property as requested.

read point-by-point responses

  1. Referee: The central claim that the bridge score is a balancing score for the mediator stage of sequential ignorability (and thereby permits a sharp pointwise envelope on the confounding function) rests on the sufficiency of the pair of treatment-specific mediator densities at a fixed mediator value. A concrete verification of this dimension-reduction step—e.g., showing that the conditional independence statements implied by sequential ignorability are preserved after conditioning on the bridge score—would strengthen the load-bearing derivation.

    Authors: We appreciate the referee highlighting the need for a more explicit verification of the dimension-reduction step. The original manuscript establishes the balancing property in Theorem 1 by constructing the bridge score from the pair of treatment-specific mediator densities evaluated at the observed mediator value and showing that this low-dimensional vector is sufficient for the mediator stage of sequential ignorability. To address the comment directly, the revised manuscript now includes an expanded proof in Section 3.1 that walks through the conditional independence statements step by step: we verify that, conditional on the bridge score and treatment, the mediator is independent of the potential outcomes (Y(1,M(0)), Y(0,M(1)), etc.) in the sense required by sequential ignorability. This is shown by demonstrating that the bridge score captures all relevant information from the mediator density ratio, preserving the ignorability condition without loss of information. The added steps make the sufficiency argument fully transparent while leaving the subsequent sharp pointwise envelope and calibration procedures unchanged. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The derivation introduces the bridge score directly from the observable treatment-specific mediator densities at a fixed value and proves its balancing property for the mediator stage of sequential ignorability via standard conditional independence arguments. The subsequent sharp pointwise envelope on the mediator-outcome confounding function is then obtained by explicit maximization over the two latent parameters conditional on that score. Neither step reduces to a fitted quantity renamed as a prediction, a self-citation chain, or an ansatz smuggled from prior work; the construction remains self-contained with independent mathematical content.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

Only the abstract is available, so the ledger is necessarily incomplete; the central construction rests on the domain assumption that sequential ignorability cannot be verified and on the newly introduced bridge score entity.

axioms (1)
  • domain assumption Mediator stage of sequential ignorability cannot be empirically verified and therefore requires explicit sensitivity analysis.
    Explicitly stated in the abstract as the motivation for the entire sensitivity framework.
invented entities (1)
  • bridge score no independent evidence
    purpose: Low-dimensional balancing score constructed from treatment-specific mediator densities
    Newly defined in the paper to enable conditioning for the sensitivity bounds.

pith-pipeline@v0.9.0 · 5728 in / 1315 out tokens · 45478 ms · 2026-05-20T08:03:38.928699+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

142 extracted references · 142 canonical work pages

  1. [1]

    Tchetgen Tchetgen , title =

    Wang Miao and Xu Shi and Yilin Li and Eric J. Tchetgen Tchetgen , title =. Statistical Theory and Related Fields , volume =. 2024 , publisher =. doi:10.1080/24754269.2024.2390748 , URL =

    work page doi:10.1080/24754269.2024.2390748 2024

  2. [2]

    Chen et al

    A flexible Bayesian g-formula for causal survival analyses with time-dependent confounding: X. Chen et al. , author=. Lifetime Data Analysis , volume=. 2025 , publisher=

    work page 2025

  3. [3]

    Improved sensitivity bounds for mediation under unmeasured mediator–outcome confounding , title =

    Arvid Sjölander and Ingeborg Waernbaum , pages =. Improved sensitivity bounds for mediation under unmeasured mediator–outcome confounding , title =. Journal of Causal Inference , doi =. 2024 , lastchecked =

    work page 2024

  4. [4]

    Epidemiology , volume=

    Mediational E-values: approximate sensitivity analysis for unmeasured mediator--outcome confounding , author=. Epidemiology , volume=. 2019 , publisher=

    work page 2019

  5. [5]

    Bias Formulas for Estimating Direct and Indirect Effects When Unmeasured Confounding Is Present , urldate =

    Saskia le Cessie , journal =. Bias Formulas for Estimating Direct and Indirect Effects When Unmeasured Confounding Is Present , urldate =

    work page

  6. [6]

    Statistics in Medicine , volume=

    Bayesian sensitivity analysis for unmeasured confounding in causal mediation analysis , author=. Statistics in Medicine , volume=. 2017 , publisher=

    work page 2017

  7. [7]

    Statistics in Medicine , volume =

    Guo, Yi and Chen, Dan and Xu, Xinming and Zhang, Zhicheng and Wen, Yu and Zheng, Xueying and Wu, Zhenyu and Qin, Xu , title =. Statistics in Medicine , volume =. doi:https://doi.org/10.1002/sim.70548 , url =. https://onlinelibrary.wiley.com/doi/pdf/10.1002/sim.70548 , note =

    work page doi:10.1002/sim.70548

  8. [8]

    2026 , eprint=

    Identification and estimation of causal mechanisms in cluster-randomized trials with post-treatment confounding using Bayesian nonparametrics , author=. 2026 , eprint=

    work page 2026

  9. [9]

    Biometrics , volume =

    Cheng, Chao and Li, Fan , title =. Biometrics , volume =. 2026 , month =. doi:10.1093/biomtc/ujag017 , url =

    work page doi:10.1093/biomtc/ujag017 2026

  10. [10]

    and Elder, Charles R

    DeBar, Lynn and Mayhew, Meghan and Benes, Lindsay and Bonifay, Allison and Deyo, Richard A. and Elder, Charles R. and Keefe, Francis J. and Leo, Michael C. and McMullen, Carmit and Owen-Smith, Ashli and Smith, David H. and Trinacty, Connie M. and Vollmer, William M. , title =. Annals of Internal Medicine , year =

    work page

  11. [11]

    Melnick, Edward R and Nath, Bidisha and Dziura, James D and Casey, Martin F and Jeffery, Molly M and Paek, Hyung and Soares, William E and Hoppe, Jason A and Rajeevan, Haseena and Li, Fangyong and Skains, Rachel M and Walter, Lauren A and Patel, Mehul D and Chari, Srihari V and Platts-Mills, Timothy F and Hess, Erik P and D. User centered clinical decisio...

    work page 2022

  12. [12]

    and Rubin, Donald B

    Rosenbaum, Paul R. and Rubin, Donald B. , title =. Biometrika , volume =. 1983 , month =. doi:10.1093/biomet/70.1.41 , url =

    work page doi:10.1093/biomet/70.1.41 1983

  13. [13]

    , title =

    Ding, Peng and Vanderweele, Tyler J. , title =. Biometrika , volume =. 2016 , month =. doi:10.1093/biomet/asw012 , url =

    work page doi:10.1093/biomet/asw012 2016

  14. [14]

    Biometrics , volume =

    Cheng, Chao and Hu, Liangyuan and Li, Fan , title =. Biometrics , volume =. 2024 , month =. doi:10.1093/biomtc/ujae045 , url =

    work page doi:10.1093/biomtc/ujae045 2024

  15. [15]

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

    Ding, Peng and Lu, Jiannan , title =. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume =. 2016 , month =. doi:10.1111/rssb.12191 , url =

    work page doi:10.1111/rssb.12191 2016

  16. [16]

    and Li, Xiaochun , title =

    Li, Lingling and Shen, Changyu and Wu, Ann C. and Li, Xiaochun , title =. American Journal of Epidemiology , volume =. 2011 , month =. doi:10.1093/aje/kwr096 , url =

    work page doi:10.1093/aje/kwr096 2011

  17. [17]

    The Annals of Applied Statistics , number =

    Liangyuan Hu and Jungang Zou and Chenyang Gu and Jiayi Ji and Michael Lopez and Minal Kale , title =. The Annals of Applied Statistics , number =. 2022 , doi =

    work page 2022

  18. [18]

    Tchetgen Tchetgen and Ilya Shpitser , title =

    Eric J. Tchetgen Tchetgen and Ilya Shpitser , title =. The Annals of Statistics , number =. 2012 , doi =

    work page 2012

  19. [19]

    and Linero, Antonio and Roy, Jason , year=

    Daniels, Michael J. and Linero, Antonio and Roy, Jason , year=. Bayesian Nonparametrics for Causal Inference and Missing Data , doi=

    work page

  20. [20]

    Journal of the American Statistical Association , volume =

    Feng Liang and Rui Paulo and German Molina and Merlise A Clyde and Jim O Berger , title =. Journal of the American Statistical Association , volume =. 2008 , publisher =. doi:10.1198/016214507000001337 , URL =

    work page doi:10.1198/016214507000001337 2008

  21. [21]

    and Foster, Dean P

    George, EdwardI. and Foster, Dean P. , title =. Biometrika , volume =. 2000 , month =. doi:10.1093/biomet/87.4.731 , url =

    work page doi:10.1093/biomet/87.4.731 2000

  22. [22]

    and Dominici, Francesca and Wang, Yun , title =

    Zigler, Corwin M. and Dominici, Francesca and Wang, Yun , title =. Biostatistics , volume =. 2012 , month =. doi:10.1093/biostatistics/kxr052 , url =

    work page doi:10.1093/biostatistics/kxr052 2012

  23. [23]

    Electronic Journal of Statistics , publisher =

    Guido Masarotto and Cristiano Varin , title =. Electronic Journal of Statistics , publisher =. 2012 , doi =

    work page 2012

  24. [24]

    and Li, Fan and Gallis, John A

    Turner, Elizabeth L. and Li, Fan and Gallis, John A. and Prague, Melanie and Murray, David M. , title =. American Journal of Public Health , volume =. 2017 , doi =. https://doi.org/10.2105/AJPH.2017.303706 , abstract =

    work page doi:10.2105/ajph.2017.303706 2017

  25. [25]

    Journal of Machine Learning Research , year =

    Yuki Ohnishi and Bikram Karmakar and Arman Sabbaghi , title =. Journal of Machine Learning Research , year =

    work page

  26. [26]

    Dunson and Antonio Lijoi and Igor Pr

    Federico Camerlenghi and David B. Dunson and Antonio Lijoi and Igor Pr. Bayesian Analysis , number =. 2019 , doi =

    work page 2019

  27. [27]

    Probability Theory and Related Fields , author =

    Pitman, Jim , title=. Probability Theory and Related Fields , year=. doi:10.1007/BF01213386 , url=

    work page doi:10.1007/bf01213386

  28. [28]

    The Annals of Statistics , number =

    Federico Camerlenghi and Antonio Lijoi and Peter Orbanz and Igor Pr. The Annals of Statistics , number =. 2019 , doi =

    work page 2019

  29. [29]

    Journal of the American Statistical Association , volume =

    Francesco Denti and Federico Camerlenghi and Michele Guindani and Antonietta Mira , title =. Journal of the American Statistical Association , volume =. 2023 , publisher =. doi:10.1080/01621459.2021.1933499 , note =

    work page doi:10.1080/01621459.2021.1933499 2023

  30. [30]

    Journal of the American Statistical Association , volume =

    Yuki Ohnishi and Fan Li , title =. Journal of the American Statistical Association , volume =. 2025 , publisher =. doi:10.1080/01621459.2025.2544366 , URL =

    work page doi:10.1080/01621459.2025.2544366 2025

  31. [31]

    Bayesian inference and decision techniques: Essays in Honor of Bruno de Finetti , editor =

    Zellner, Arnold , title =. Bayesian inference and decision techniques: Essays in Honor of Bruno de Finetti , editor =. 1986 , series =

    work page 1986

  32. [32]

    Biometrics , volume =

    Bae, Woojung and Daniels, Michael J and Perri, Michael G , title =. Biometrics , volume =. 2024 , month =. doi:10.1093/biomtc/ujae099 , url =

    work page doi:10.1093/biomtc/ujae099 2024

  33. [33]

    and Williams, Nicholas and Díaz, Iván , title =

    Rudolph, Kara E. and Williams, Nicholas and Díaz, Iván , title =. Biometrics , volume =. doi:https://doi.org/10.1111/biom.13850 , url =. https://onlinelibrary.wiley.com/doi/pdf/10.1111/biom.13850 , abstract =

    work page doi:10.1111/biom.13850

  34. [34]

    Biometrics , volume =

    Hong, Guanglei and Yang, Fan and Qin, Xu , title =. Biometrics , volume =. doi:https://doi.org/10.1111/biom.13705 , url =. https://onlinelibrary.wiley.com/doi/pdf/10.1111/biom.13705 , abstract =

    work page doi:10.1111/biom.13705

  35. [35]

    2023 , eprint=

    Truncation Approximation for Enriched Dirichlet Process Mixture Models , author=. 2023 , eprint=

    work page 2023

  36. [36]

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

    Cheng, Chao and Li, Fan , title =. Journal of the Royal Statistical Society Series B: Statistical Methodology , year =. doi:10.1093/jrsssb/qkaf037 , url =

    work page doi:10.1093/jrsssb/qkaf037

  37. [37]

    Bayesian Analysis , number =

    Sara Wade and Silvia Mongelluzzo and Sonia Petrone , title =. Bayesian Analysis , number =. 2011 , doi =

    work page 2011

  38. [38]

    Dunson and Sonia Petrone and Lorenzo Trippa , title =

    Sara Wade and David B. Dunson and Sonia Petrone and Lorenzo Trippa , title =. Journal of Machine Learning Research , year =

    work page

  39. [39]

    How Effective are Cash Transfers at Improving Nutritional Status? , journal =

    James Manley and Seth Gitter and Vanya Slavchevska , keywords =. How Effective are Cash Transfers at Improving Nutritional Status? , journal =. 2013 , issn =. doi:https://doi.org/10.1016/j.worlddev.2013.03.010 , url =

    work page doi:10.1016/j.worlddev.2013.03.010 2013

  40. [40]

    Fonctions de repartition an dimensions et leurs marges , author=. Publ. inst. statist. univ. Paris , volume=

    work page

  41. [41]

    Bayesian Analysis , number =

    Andr. Bayesian Analysis , number =. 2012 , doi =

    work page 2012

  42. [42]

    Eddy , title =

    Seymour Geisser and William F. Eddy , title =. Journal of the American Statistical Association , volume =. 1979 , publisher =

    work page 1979

  43. [43]

    A. E. Gelfand and D. K. Dey , journal =. Bayesian Model Choice: Asymptotics and Exact Calculations , urldate =

    work page

  44. [44]

    Nonparametric inference for interventional effects with multiple mediators , title =

    David Benkeser and Jialu Ran , pages =. Nonparametric inference for interventional effects with multiple mediators , title =. Journal of Causal Inference , doi =. 2021 , lastchecked =

    work page 2021

  45. [45]

    , title=

    Vansteelandt, Stijn and Daniel, Rhian M. , title=. Epidemiology , year=

    work page

  46. [46]

    and Vansteelandt, Stijn and Robins, James M

    VanderWeele, Tyler J. and Vansteelandt, Stijn and Robins, James M. , address =. Effect Decomposition in the Presence of an Exposure-Induced Mediator-Outcome Confounder , volume =. Epidemiology (Cambridge, Mass.) , keywords =. 2014 , abstract =

    work page 2014

  47. [47]

    Aronow and Cyrus Samii , title =

    Peter M. Aronow and Cyrus Samii , title =. The Annals of Applied Statistics , number =. 2017 , doi =

    work page 2017

  48. [48]

    Statistical Methods in Medical Research , volume=

    Demystifying estimands in cluster-randomised trials , author=. Statistical Methods in Medical Research , volume=. 2024 , publisher=

    work page 2024

  49. [49]

    A Bayesian analysis of the multinomial probit model using marginal data augmentation , journal =

    Kosuke Imai and David A. A Bayesian analysis of the multinomial probit model using marginal data augmentation , journal =. 2005 , issn =. doi:https://doi.org/10.1016/j.jeconom.2004.02.002 , url =

    work page doi:10.1016/j.jeconom.2004.02.002 2005

  50. [50]

    American Journal of Epidemiology , volume =

    Charters, Thomas J and Kaufman, Jay S and Nandi, Arijit , title = ". American Journal of Epidemiology , volume =. 2023 , month =. doi:10.1093/aje/kwac165 , url =

    work page doi:10.1093/aje/kwac165 2023

  51. [51]

    Albert and Siddhartha Chib , title =

    James H. Albert and Siddhartha Chib , title =. Journal of the American Statistical Association , volume =. 1993 , publisher =. doi:10.1080/01621459.1993.10476321 , URL =

    work page doi:10.1080/01621459.1993.10476321 1993

  52. [52]

    Biometrika , volume =

    Chib, Siddhartha and Greenberg, Edward , title = ". Biometrika , volume =. 1998 , month =. doi:10.1093/biomet/85.2.347 , url =

    work page doi:10.1093/biomet/85.2.347 1998

  53. [53]

    Journal of the American Statistical Association , volume =

    Abel Rodríguez and David B Dunson and Alan E Gelfand , title =. Journal of the American Statistical Association , volume =. 2008 , publisher =

    work page 2008

  54. [54]

    Dunson and Abel Rodr

    David B. Dunson and Abel Rodr. Bayesian Analysis , number =. 2011 , doi =

    work page 2011

  55. [55]

    and Park, Ju-Hyun , title = "

    Dunson, David B. and Park, Ju-Hyun , title = ". Biometrika , volume =. 2008 , month =. doi:10.1093/biomet/asn012 , url =

    work page doi:10.1093/biomet/asn012 2008

  56. [56]

    Ogburn and Oleg Sofrygin and Iván Díaz and Mark J

    Elizabeth L. Ogburn and Oleg Sofrygin and Iván Díaz and Mark J. van der Laan , title =. Journal of the American Statistical Association , volume =. 2024 , publisher =. doi:10.1080/01621459.2022.2131557 , URL =

    work page doi:10.1080/01621459.2022.2131557 2024

  57. [57]

    A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , journal =

    James Robins , abstract =. A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , journal =. 1986 , issn =. doi:https://doi.org/10.1016/0270-0255(86)90088-6 , url =

    work page doi:10.1016/0270-0255(86)90088-6 1986

  58. [58]

    Statistical Science , number =

    Kosuke Imai and Luke Keele and Teppei Yamamoto , title =. Statistical Science , number =. 2010 , doi =

    work page 2010

  59. [59]

    doi:10.1080/01621459.2023

    Bingkai Wang and Chan Park and Dylan S. Small and Fan Li , title =. Journal of the American Statistical Association , volume =. 2024 , publisher =. doi:10.1080/01621459.2023.2289693 , URL =

    work page doi:10.1080/01621459.2023.2289693 2024

  60. [60]

    and Roy, Jason A

    Daniels, Michael J. and Roy, Jason A. and Kim, Chanmin and Hogan, Joseph W. and Perri, Michael G. Bayesian inference for the causal effect of mediation. Biometrics

    work page

  61. [61]

    Statistics in Medicine , volume =

    Kim, Chanmin and Daniels, Michael and Li, Yisheng and Milbury, Kathrin and Cohen, Lorenzo , title =. Statistics in Medicine , volume =. doi:https://doi.org/10.1002/sim.7572 , url =. https://onlinelibrary.wiley.com/doi/pdf/10.1002/sim.7572 , abstract =

    work page doi:10.1002/sim.7572

  62. [62]

    2024 , eprint=

    Semiparametric causal mediation analysis in cluster-randomized experiments , author=. 2024 , eprint=

    work page 2024

  63. [63]

    Bayesian Analysis , number =

    Yuki Ohnishi and Arman Sabbaghi , title =. Bayesian Analysis , number =. 2024 , doi =

    work page 2024

  64. [64]

    Hudgens and M

    Michael G. Hudgens and M. Elizabeth Halloran , doi =. Toward causal inference with interference , volume =. Journal of the American Statistical Association , keywords =

    work page

  65. [65]

    Multivariate Behavioral Research , volume =

    Soojin Park and David Kaplan , title =. Multivariate Behavioral Research , volume =. 2015 , publisher =. doi:10.1080/00273171.2014.1003770 , note =

    work page doi:10.1080/00273171.2014.1003770 2015

  66. [66]

    Biostatistics , volume =

    Roy, Samrat and Daniels, Michael J and Roy, Jason , title = ". Biostatistics , volume =. 2024 , month =. doi:10.1093/biostatistics/kxad038 , url =

    work page doi:10.1093/biostatistics/kxad038 2024

  67. [67]

    Multilevel Mechanisms of Implementation Strategies in Mental Health: Integrating Theory, Research, and Practice

    Williams, Nathaniel J. Multilevel Mechanisms of Implementation Strategies in Mental Health: Integrating Theory, Research, and Practice. Administration and Policy in Mental Health and Mental Health Services Research

    work page

  68. [68]

    Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence , pages =

    Pearl, Judea , title =. Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence , pages =. 2001 , isbn =

    work page 2001

  69. [69]

    Identifiability and exchangeability for direct and indirect effects

    Robins, J M and Greenland, S. Identifiability and exchangeability for direct and indirect effects. Epidemiology

    work page

  70. [70]

    Reich and Montserrat Fuentes , title =

    Brian J. Reich and Montserrat Fuentes , title =. The Annals of Applied Statistics , number =. 2007 , doi =

    work page 2007

  71. [71]

    Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences , volume =

    Li, Fan and Ding, Peng and Mealli, Fabrizia , title =. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences , volume =. 2023 , doi =. https://royalsocietypublishing.org/doi/pdf/10.1098/rsta.2022.0153 , abstract =

    work page doi:10.1098/rsta.2022.0153 2023

  72. [72]

    Journal of the American Statistical Association , author =

    Man-Wai Ho and Wanzhu Tu and Pulak Ghosh and Ram C. Tiwari , title =. Journal of the American Statistical Association , volume =. 2013 , publisher =. doi:10.1080/01621459.2012.734164 , URL =

    work page doi:10.1080/01621459.2012.734164 2013

  73. [73]

    A CONSTRUCTIVE DEFINITION OF DIRICHLET PRIORS , urldate =

    Jayaram Sethuraman , journal =. A CONSTRUCTIVE DEFINITION OF DIRICHLET PRIORS , urldate =

    work page

  74. [74]

    Bayesian dynamic density estimation , volume =

    Abel Rodr\'iguez and Enrique ter Horst , doi =. Bayesian dynamic density estimation , volume =. Bayesian Analysis , keywords =

    work page

  75. [75]

    Generalized Spatial Dirichlet Process Models , volume =

    Jason A Duan and Michele Guindani and Alan E Gelfand , doi =. Generalized Spatial Dirichlet Process Models , volume =. Biometrika , pages =

    work page

  76. [76]

    and Müller, Peter , title = "

    De la Cruz-Mesía, Rolando and Quintana, Fernando A. and Müller, Peter , title = ". Journal of the Royal Statistical Society Series C: Applied Statistics , volume =. 2007 , month =. doi:10.1111/j.1467-9876.2007.00569.x , url =

    work page doi:10.1111/j.1467-9876.2007.00569.x 2007

  77. [77]

    Gelfand and Athanasios Kottas and Steven N

    Alan E. Gelfand and Athanasios Kottas and Steven N. MacEachern , doi =. Bayesian nonparametric spatial modeling with dirichlet process mixing , volume =. Journal of the American Statistical Association , keywords =

    work page

  78. [78]

    Rosner and Steven N

    Maria De Iorio and Peter Müller and Gary L. Rosner and Steven N. MacEachern , doi =. An anova model for dependent random measures , volume =. Journal of the American Statistical Association , keywords =

    work page

  79. [79]

    The Annals of Applied Statistics , number =

    Alex Diana and Eleni Matechou and Jim Griffin and Alison Johnston , title =. The Annals of Applied Statistics , number =. 2020 , doi =

    work page 2020

  80. [80]

    2024 , eprint=

    Covariate-dependent hierarchical Dirichlet process , author=. 2024 , eprint=

    work page 2024

Showing first 80 references.