Pith. sign in

REVIEW 4 major objections 6 minor 61 references

Bayesian Transfer Learning for Enhanced Estimation and Inference

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read TRADER, a source-guided horseshoe prior, lets Bayesian transfer learning borrow from multiple pre-trained models and provably improves estimation and coverage in high-dimensional regression when sources align with the target.

desk verdict A promising transfer-learning prior with a load-bearing bug: the Dirichlet weights are undefined for negative or zero cosine similarities, which is the very regime the paper claims to handle. read the letter →

arxiv 2412.02986 v2 pith:OWNO3ZGC submitted 2024-12-04 stat.ME

Daoyuan Lai , Oscar Hernan Madrid Padilla , Xiang Li , Tian Gu This is my paper · ORCID
classification stat.ME MSC 62F1562J0762G20
keywords Bayesiantransferlearninghorseshoepriorhigh-dimensionallinearregressionposteriorcontractionuncertaintyquantificationmulti-sourcecoverageprobabilitysummary-leveldata
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper introduces TRADER, a Bayesian transfer learning method that shrinks the target regression coefficients toward a weighted average of rescaled estimates from several pre-trained source models. The authors aim to show that when at least one source is informative, the posterior contracts around the true coefficients at a faster rate than the standard horseshoe prior achieves using target data alone, and that this also repairs the well-known undercoverage of moderate-strength signals. If correct, this would give practitioners a communication-efficient and privacy-preserving way to borrow information across studies while still producing credible intervals with frequentist validity. The method needs only the source parameter estimates, not the underlying individual-level data.

What carries the argument

The source-guided horseshoe prior. Each target coefficient $\beta_j$ is given a normal prior with mean $\eta_{K+1}\cdot 0$ plus the weighted sum of rescaled source estimates, where source $k$ is scaled by the ratio $\|\hat\beta_{\text{val}}\|/\|\hat\omega^{(k)}\|$ to match the target's scale and the weights $(\eta_1,\ldots,\eta_{K+1})$ follow a Dirichlet distribution whose concentration parameters are cosine similarities between each source estimate and the target validation estimate. The global shrinkage parameter $\tau$ is set not by a half-Cauchy hyperprior but by the informative-element count $\psi=\sum_j\kappa_j$, which determines how much the posterior leans on the source; this avoids the instability of $\tau$ collapsing to zero when sources are near the truth.

What would settle it

Run a simulation where the validation split is small (e.g., $n_0/10$) and the source estimates are noisy or biased, so that the estimated scaling factors and cosine-similarity weights make the combined source estimate violate the sparse-support condition (6); if the posterior contraction rate then matches the target-only rate or the credible intervals undercover for moderate signals, the claim that the adaptive construction preserves the theoretical guarantees is refuted.

Watch

Extended reading notes

Core claim

The central claim is that a horseshoe prior whose mean is a data-adaptive combination of rescaled source estimates enjoys posterior contraction rates up to $\sqrt{\tilde s(r)\log p/n_0}$ with $\tilde s(r)$ potentially much smaller than the true sparsity $s$ when the combined source estimate is well aligned, while degenerating gracefully to the target-only rate $\sqrt{s\log p/n_0}$ when no source is helpful. Theorems 1 and 2 establish this as an oracle inequality with the effective sparsity determined by how much the combined source estimate deviates from the truth, and Theorem 3 shows that the marginal posterior for each coefficient can be approximated by a normal centered near the MLE with a bias term that shrinks at the improved rate, which is what restores nominal coverage for coefficients of moderate signal strength. The paper argues, with simulations and a Hispanic subgroup analysis of insulin and HbA1C, that this yields real gains in estimation and inference over using the target data alone and over an individual-level-data transfer method that is sensitive to scale mismatches.

Load-bearing premise

The theory treats the combined source estimate and oracle weights as fixed and independent of the target data, but the method constructs them from a validation split of the target data; the load-bearing premise is that this data-driven construction still satisfies the support-size and alignment conditions (such as condition (6)) that the theorems require.

Editorial extensions

If this is right

  • If at least one source estimate is sparse and close to the target, the effective sparsity $\tilde s(r)$ is smaller than $s$, so credible intervals shrink faster than target-only horseshoe and still cover the truth.
  • With multiple sources that individually capture only part of the support, the adaptive Dirichlet weighting recovers an oracle-weighted combination, improving the contraction rate without knowing which sources are informative.
  • In the worst case of uninformative sources, the prior collapses to a standard zero-mean horseshoe, so TRADER cannot perform worse than target-only analysis.
  • The real-data example suggests that borrowing from larger racial/ethnic groups can turn a non-significant association in a small underrepresented subgroup into a significant, appropriately uncertain estimate.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The validation-split estimation of the scaling factors and weights is not covered by the theory, so in practice the guarantees require the validation estimate to be sufficiently accurate; users should check the stability of the constructed source estimate.
  • The mechanism suggests a recipe for other shrinkage priors: any global-local prior can be 'guided' by replacing the zero mean with an adaptively weighted external estimate, potentially extending the result to Dirichlet–Laplace or horseshoe+ priors.
  • Because TRADER needs only summary statistics, it could be combined with federated learning pipelines where each site runs its own horseshoe regression and shares only the fitted coefficients.
  • A testable extension would be to let the Dirichlet concentration parameters themselves be estimated from the data rather than fixed as the raw cosine similarities, which could improve robustness in small-validation settings.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes TRADER, a Bayesian transfer learning method for high-dimensional linear regression in which multiple source datasets are used only through pre-trained coefficient estimates. The prior centers each target coefficient on a rescaled, adaptively weighted average of source estimates, with Dirichlet weights informed by cosine similarities to a validation estimate and with horseshoe-type local-global shrinkage. The manuscript claims oracle posterior contraction rates faster than target-only horseshoe when sources are well aligned, protection against negative transfer from heterogeneous sources, and improved finite-sample frequentist coverage for moderate signal strengths. Theoretical results include an oracle contraction inequality (Theorems 1-2), a heuristic for the global shrinkage parameter (Proposition 1), and a finite-sample marginal posterior approximation (Theorem 3). The paper also presents simulation comparisons with target-only horseshoe and TransGLM, plus a MIMIC-III application estimating the association between HbA1C and insulin use in Hispanic patients.

Significance. If the central claims were fully established, TRADER would address a timely and practical gap: multi-source transfer learning using only summary-level source estimates, with communication efficiency and privacy preservation, together with uncertainty quantification. The paper's strengths are its clear problem framing, the explicit oracle-type contraction rates in terms of the effective sparsity of the source-target difference, the worst-case robustness statement in Corollary 2, and the broad simulation study covering scale and correlation heterogeneity. However, several load-bearing points are not yet resolved: the Dirichlet prior as written is not a valid distribution for non-positive cosine similarities, the estimator bβ_val used to build the prior is not defined in the high-dimensional regimes considered, the theory treats the source vector and weights as fixed and independent of target data while the method estimates them from a validation split, and Theorem 3's 'there exists F_j' construction does not directly establish a property of the actual posterior. These issues block acceptance of the core methodological and coverage claims in the current version.

major comments (4)
  1. [Section 2.2, Eq. (2)] The Dirichlet prior on the source weights is not a well-defined probability distribution when any cosine similarity θ_k is non-positive. The Dirichlet density requires strictly positive concentration parameters, but θ_k = bω^(k)T bβ_val / (||bω^(k)|| ||bβ_val||) can be negative for anti-correlated sources and zero for orthogonal sources. The paper explicitly targets heterogeneous and uninformative sources, and Simulation Settings II and III generate sources with near-zero or small positive correlations. If the implementation clips, shifts, or exponentiates θ_k, that transformation is absent from the manuscript. As written, the prior in Eq. (2) is internally inconsistent in exactly the regime the method claims to handle, and the simulations are not reproducible without an additional definition.
  2. [Section 2.1, Step 1] The paper does not specify how bβ_val is computed, and in the settings considered it cannot be the usual OLS estimator. In the simulations n0 = 120 and p = 200, so a 1/3 validation split gives n_val = 40 < p, making (X_val^T X_val)^{-1} undefined. The same issue appears in the real-data application with p = 121 and a Hispanic target sample of n = 189, giving n_val ≈ 63 < p. Since bβ_val enters both the scaling factors ||bβ_val||/||bω^(k)|| and the cosine similarities θ_k, the proposed prior is not fully specified in the regimes where the method is evaluated. An explicit regularized estimator or a Bayesian procedure for bβ_val is needed.
  3. [Sections 2.1-2.2 and Theorems 1-2] The theoretical results treat the scaled source vector eω and the source weights η* as fixed and independent of the target data, while the method estimates them from a validation split of the target data. Theorem 2 explicitly assumes that eω^(1),...,eω^(K) are independent of the target data, but the scaling factors ||bβ_val||/||bω^(k)|| depend on target validation data, and the Dirichlet parameters are functions of bβ_val. No theorem or argument shows that the data-driven eω and η satisfy conditions such as (6) or the existence of the oracle weights η*. In addition, Theorems 1 and 2 require τ of order p^{-u} or n0^{-u} for an unspecified 'appropriate u', while Proposition 1 fixes a specific τ0 = (p - bψ)/(√n0 bψ); the paper does not verify that τ0 satisfies the theorem conditions. This gap separates the oracle statements from the actual TRADER procedure.
  4. [Theorem 3, Eqs. (12)-(13)] The statement 'there exists a distribution F_j with support contained in B' does not establish a property of the actual marginal posterior π(β_j | D0, σ²). The theorem constructs a marginal f_j from an arbitrary F_j and bounds its total variation distance to the actual posterior, but unless the actual posterior of β_{-j} is shown to concentrate on B, the bound is not informative about the real posterior. No lemma in the main text proves posterior mass of B. Part (ii) has the same problem, and its conditional density involving an unnormalized function h(β_j) and an indicator I(β_j ∈ R(β_{-j})) is not shown to be a valid probability kernel. Since Theorem 3 is the basis for the paper's finite-sample coverage claims, this is a load-bearing gap.
minor comments (6)
  1. [Section 4.1] The figure reference appears as 'Figure ??' and should be corrected to the appropriate figure number.
  2. [Table 1] The Caucasian-only 95% confidence interval [-1.349, -0.132] is not centered at the reported point estimate -1.091; please verify the reported interval or point estimate.
  3. [Section 2.4, Proposition 1] The description of ψ as 'counting' the number of informative elements is informal because κ_j are continuous shrinkage factors; the formula E(ψ | τ) = p/(1 + τ√n0) should be derived explicitly using the half-Cauchy marginal calculation and the approximation X^(0)T X^(0) ≈ n0 I.
  4. [Assumption 3 and Theorem 3] The symbol p is used both for the dimension and for a separate threshold integer (e.g., 'there exists p > s, p ≍ s' and 'p > max{e_s(r_sup), s}'); a different symbol such as \bar p would remove the ambiguity.
  5. [Section 4, MSE definition] In the definition MSE = (1/p) Σ_{j=1}^p (bβ_mean_j - β_i)^2, the subscript i in β_i should be j.
  6. [Sections 2.1 and 2.4] No sensitivity analysis is reported for the user-chosen hyperparameters: the 1/3 validation split fraction, bψ = p/2, and the Dirichlet concentration ζ = 1; given the role of τ and the validation estimate in source weighting, some sensitivity checks would strengthen the empirical claims.

Circularity Check

1 steps flagged · score 5.0 of 10

Theorem 3's coverage guarantee is constructed via an arbitrary auxiliary distribution F_j and a chosen interval R, making the finite-sample coverage claim partly definitional; the contraction-rate theory remains conditional and non-circular.

  1. self definitional [Theorem 3, Section 3, Eqs. (11)-(13)]
    "(i) Suppose |eωj − β∗ j | > Cσϵn for some constant C > 0. Then, there exists a distribution Fj with support contained in B such that the random variable βj generated as βj | β−j ∼ N(gj(β−j), σ2 j ), β−j ∼ Fj, has a marginal distribution fj that satisfies ∥fj − π(βj | D0, σ2)∥TV = O(|β∗ j − eωj|2/σp2 + exp(−c2n0ϵ2n) + σϵn/|β∗ j − eωj|)."

    The theorem does not analyze the posterior induced by model (1) and prior (2). Equation (9) states the actual full conditional is N(gj(β−j), σ2 j ) · πβj (βj); Theorem 3 instead replaces the prior factor by an arbitrary h(βj) (or omits it) and postulates an auxiliary F_j with support in B, a set defined using the true β*. The interval R(β−j) is then defined so that eωj lies inside it and h is chosen to blow up at eωj, so concentration near the source is true by construction. Remark 5 and the abstract interpret this existence statement as "TRADER achieves desired frequentist coverage probabilities". The coverage conclusion is therefore assumed into existence rather than derived from the actual marginal posterior, which is a self-definitional reduction of the central inference claim.

full rationale

The main contraction-rate results (Theorems 1 and 2) are conditional oracle inequalities: they take rescaled source estimates eω and oracle weights η* as fixed external inputs satisfying sparsity conditions, and derive posterior concentration about β*. The rate formula is not identical to the assumptions and the bounds are not fitted from the target outcome, so those results are not circular. The paper does not supply a theorem covering the data-driven construction of η from a target validation split in Section 2.2, but that is a support gap rather than a definitional equivalence and is treated as such here. The clear circular step is Theorem 3: the claimed finite-sample marginal posterior behavior is an existence statement about an auxiliary distribution F_j, with an interval R and factor h engineered so that the desired conclusion holds by construction; the actual full conditional in Eq. (9) differs from the one used in the theorem. The abstract's coverage claim therefore rests in part on this constructed object. Separately, there are non-circular correctness issues — e.g., the Dirichlet concentration parameters in Eq. (2) can be non-positive for negatively correlated or orthogonal sources — but these do not affect the circularity score. Overall, the estimation-rate contribution has independent content, while the inference guarantee is partially circular, giving a score of 5.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the validity of the data-driven prior construction, which is not covered by the theoretical assumptions. The rescaling and weighting use target validation data, breaking the fixed-source independence assumed in the theorems. The prior guess bψ and validation split are hand-chosen hyperparameters.

free parameters (3)
  • bψ (prior number of informative elements) = p/2
    Chosen by hand to set τ0 via Proposition 1, reflecting a prior belief that 50% of elements borrow from sources. No sensitivity analysis is provided.
  • Validation split fraction = 1/3
    Fraction of target data reserved to compute bβ_val for rescaling and cosine similarity weights. This affects the quality of source alignment and is not covered by theory.
  • Dirichlet concentration ζ = 1
    Fixed concentration for the weight prior. No tuning or sensitivity analysis is reported.
assumptions (5)
  • domain assumption Covariates uniformly bounded, x_ij ∈ [-1,1] (Assumption 1)
    Used for the theoretical results; may not hold in all applications.
  • standard math Sparsity condition s log p << n0 (Assumption 2)
    Standard high-dimensional sparsity assumption.
  • standard math Restricted eigenvalue condition λ_min(X_ξ^T X_ξ) ≥ n0 λ0 for all |ξ| ≤ p (Assumption 3)
    Standard identifiability condition for high-dimensional regression.
  • domain assumption Source estimates eω are independent of the target data and treated as fixed (Theorems 1-2)
    The method uses validation data from the target to rescale and weight sources, so this independence is violated in practice.
  • ad hoc to paper There exist oracle weights η* satisfying condition (6) and support overlap conditions (Theorem 2, Corollary 2)
    The data-driven Dirichlet weights are not shown to satisfy these conditions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bayesian Transfer Learning for Enhanced Estimation and Inference." pith.science (2026). https://pith.science/paper/OWNO3ZGC

@misc{pith2026241202986,
  author       = {Pith},
  title        = {Pith review of: Bayesian Transfer Learning for Enhanced Estimation and Inference},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OWNO3ZGC}},
  note         = {Machine review of arXiv:2412.02986}
}
read the original abstract

Transfer learning enhances model performance in a target population with limited samples by leveraging knowledge from related studies. While many works focus on improving predictive performance, challenges of statistical inference persist. Bayesian approaches naturally offer uncertainty quantification for parameter estimates, yet existing Bayesian transfer learning methods are typically limited to single-source scenarios or require individual-level data. We introduce TRansfer leArning via guideD horseshoE prioR (TRADER), a novel approach enabling multi-source transfer through pre-trained models in high-dimensional linear regression. TRADER shrinks target parameters towards a weighted average of source estimates, accommodating sources with different scales. Theoretical investigation shows that TRADER achieves faster posterior contraction rates than standard continuous shrinkage priors when sources align well with the target while preventing negative transfer from heterogeneous sources. The analysis of finite-sample marginal posterior behavior reveals that TRADER achieves desired frequentist coverage probabilities, even for coefficients with moderate signal strength--a scenario where standard continuous shrinkage priors struggle. Extensive numerical studies and a real-data application estimating the association between blood glucose and insulin use in the Hispanic diabetic population demonstrate that TRADER improves estimation and inference accuracy over continuous shrinkage priors using target data alone, while outperforming a state-of-the-art transfer learning method that requires individual-level data.

Figures

Figures reproduced from arXiv: 2412.02986 by the authors.

Figure 1
Figure 1. Schematic illustration of an existing challenge and the proposed two-step proce [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. (Simulation Setting I) Comparison of TRADER, TransGLM, and standard horse [PITH_FULL_IMAGE:figures/full_fig_p024_2.png] view at source ↗
Figure 3
Figure 3. (Simulation setting II) Comparison of TRADER, TransGLM, and standard [PITH_FULL_IMAGE:figures/full_fig_p027_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: (Simulation setting III) Comparison of TRADER, TransGLM, and the target-only [PITH_FULL_IMAGE:figures/full_fig_p029_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

61 extracted references · 46 canonical work pages

  1. [1]

    A., Williams, J

    Abba, M. A., Williams, J. P., and Reich, B. J. (2024). A Bayesian shrinkage estimator for transfer learning . arXiv preprint arXiv:2403.17321

    arXiv 2024

  2. [2]

    and Reddy, C

    Al-Stouhi, S. and Reddy, C. K. (2016). Transfer learning for class imbalance problems with inadequate data. Knowledge and Information Systems , 48:201--228

    work page 2016

  3. [3]

    S., Stey, P., Jain, S., Biron, D

    Anand, R. S., Stey, P., Jain, S., Biron, D. R., Bhatt, H., Monteiro, K., Feller, E., Ranney, M. L., Sarkar, I. N., and Chen, E. S. (2018). Predicting mortality in diabetic ICU patients using machine learning and severity indices. American Medical Informatics Association Summits on Translational Science Proceedings , 2018:310

    work page 2018

  4. [4]

    and Ghosh, M

    Bai, R. and Ghosh, M. (2019). Large-scale multiple hypothesis testing with the normal-beta prime prior. Statistics , 53(6):1210--1233

    work page 2019

  5. [5]

    Bastani, H. (2021). Predicting with proxies: Transfer learning in high dimension. Management Science , 67(5):2964--2984

    2021

  6. [6]

    G., and Willard, B

    Bhadra, A., Datta, J., Polson, N. G., and Willard, B. (2017). The Horseshoe+ Estimator of Ultra-Sparse Signals . Bayesian Analysis , 12(4):1105 -- 1131

    work page 2017

  7. [7]

    G., and Willard, B

    Bhadra, A., Datta, J., Polson, N. G., and Willard, B. (2019). Lasso Meets Horseshoe: A Survey . Statistical Science , 34(3):405 -- 427

    work page 2019

  8. [8]

    S., and Dunson, D

    Bhattacharya, A., Pati, D., Pillai, N. S., and Dunson, D. B. (2015). Dirichlet--laplace priors for optimal shrinkage. Journal of the American Statistical Association , 110(512):1479--1490

    2015

Show all 61 references
  1. [9]

    Cai, T., Li, M., and Liu, M. (2024). Semi-supervised triply robust inductive transfer learning. Journal of the American Statistical Association , (just-accepted):1--22

    2024

  2. [10]

    Cai, T., Liu, M., and Xia, Y. (2022). Individual data protected integrative regression analysis of high-dimensional heterogeneous data. Journal of the American Statistical Association , 117(540):2105--2119

    2022

  3. [11]

    M., Polson, N

    Carvalho, C. M., Polson, N. G., and Scott, J. G. (2009). Handling sparsity via the horseshoe. In Artificial Intelligence and Statistics , pages 73--80. PMLR

    2009

  4. [12]

    M., Polson, N

    Carvalho, C. M., Polson, N. G., and Scott, J. G. (2010). The horseshoe estimator for sparse signals. Biometrika , 97(2):465--480

    2010

  5. [13]

    Chatterjee, N., Chen, Y.-H., Maas, P., and Carroll, R. J. (2016). Constrained maximum likelihood estimation for model calibration using summary-level information from external big data sources. Journal of the American Statistical Association , 111(513):107--117

    2016

  6. [14]

    Chen, E., Chen, X., Jing, W., and Zhang, Y. (2024). Distributed tensor principal component analysis. arXiv preprint arXiv:2405.11681

    2024 arXiv

  7. [15]

    M., Gu, T., Tomlins, S

    Cheng, W., Taylor, J. M., Gu, T., Tomlins, S. A., and Mukherjee, B. (2019). Informing a risk prediction model for binary outcomes with external coefficient information. Journal of the Royal Statistical Society Series C: Applied Statistics , 68(1):121--139

    2019

  8. [16]

    R., Moore, J

    Duan, R., Boland, M. R., Moore, J. H., and Chen, Y. (2018). Odal: A one-shot distributed algorithm to perform logistic regressions on electronic health records data from multiple clinical sites. In BIOCOMPUTING 2019: Proceedings of the Pacific Symposium , pages 30--41. World S...

    2018

  9. [17]

    J., Tong, J., Liang, C

    Duan, R., Luo, C., Schuemie, M. J., Tong, J., Liang, C. J., Chang, H. H., Boland, M. R., Bian, J., Xu, H., Holmes, J. H., et al. (2020). Learning from local to global: An efficient distributed algorithm for modeling time-to-event data. Journal of the American Medical Informati...

    2020

  10. [18]

    Duan, R., Ning, Y., and Chen, Y. (2022). Heterogeneity-aware and communication-efficient distributed statistical inference. Biometrika , 109(1):67--83

    2022

  11. [19]

    N., Leet, R

    Ford, C. N., Leet, R. W., Kipling, L., Rhee, M. K., Jackson, S. L., Wilson, P. W., Phillips, L. S., and Staimez, L. R. (2019). Racial differences in performance of HbA1c for the classification of diabetes and prediabetes among US adults of non-Hispanic black and white race . D...

    2019

  12. [20]

    and Zhou, X

    Gao, B. and Zhou, X. (2024). MESuSiE enables scalable and powerful multi-ancestry fine-mapping of causal variants in genome-wide association studies. Nature Genetics , 56(1):170--179

    2024

  13. [21]

    Gu, T., Han, Y., and Duan, R. (2022a). Robust angle-based transfer learning in high dimensions. arXiv preprint arXiv:2210.12759

    2022 arXiv

  14. [22]

    Gu, T., Han, Y., and Duan, R. (2022b). A transfer learning approach based on random forest with application to breast cancer prediction in underrepresented populations. In PACIFIC SYMPOSIUM ON BIOCOMPUTING 2023: Kohala Coast, Hawaii, USA, 3--7 January 2023 , pages 186--197. Wo...

    2022

  15. [23]

    H., and Duan, R

    Gu, T., Lee, P. H., and Duan, R. (2023a). COMMUTE : communication-efficient transfer learning for multi-site risk prediction. Journal of Biomedical Informatics , 137:104243

    2023

  16. [24]

    Gu, T., Li, S., and Duan, R. (2024). On the equivalence of transfer learning methods. arXiv preprint

    2024

  17. [25]

    M., Cheng, W., and Mukherjee, B

    Gu, T., Taylor, J. M., Cheng, W., and Mukherjee, B. (2019). Synthetic data method to incorporate external information into a current study. Canadian Journal of Statistics , 47(4):580--603

    2019

  18. [26]

    M., and Mukherjee, B

    Gu, T., Taylor, J. M., and Mukherjee, B. (2023b). A meta-inference framework to integrate multiple external models into a current study. Biostatistics , 24(2):406--424

    2023

  19. [27]

    Gu, T., Taylor, J. M. G., and Mukherjee, B. (2023c). A synthetic data integration framework to leverage external summary-level information from heterogeneous populations. Biometrics , 79(4):3831--3845

    2023

  20. [28]

    and Lawless, J

    Han, P. and Lawless, J. F. (2019). Empirical likelihood estimation using auxiliary summary information with different covariate distributions. Statistica Sinica , 29(3):1321--1342

    2019

  21. [29]

    M., and Mukherjee, B

    Han, P., Taylor, J. M., and Mukherjee, B. (2023). Integrating information from existing risk prediction models with no model details. Canadian Journal of Statistics , 51(2):355--374

    2023

  22. [30]

    Hector, E. C. and Martin, R. (2024). Turning the information-sharing dial: efficient inference from different data sources. Electronic Journal of Statistics , 18(2):2974--3020

    2024

  23. [31]

    P., and Hector, E

    Hickey, J., Williams, J. P., and Hector, E. C. (2024). Transfer Learning with Uncertainty Quantification: Random Effect Calibration of Source to Target (RECaST) . Journal of Machine Learning Research , 25(338):1--40

    2024

  24. [32]

    and Rao, J

    Ishwaran, H. and Rao, J. S. (2005). Spike and slab variable selection: Frequentist and Bayesian strategies . The Annals of Statistics , 33(2):730 -- 773

    2005

  25. [33]

    and Montanari, A

    Javanmard, A. and Montanari, A. (2014). Confidence intervals and hypothesis testing for high-dimensional regression. Journal of Machine Learning Research , 15(1):2869--2909

    2014

  26. [34]

    and Yin, G

    Jin, H. and Yin, G. (2021). Unit information prior for adaptive information borrowing from multiple historical datasets. Statistics in Medicine , 40(25):5657--5672

    2021

  27. [35]

    E., Pollard, T

    Johnson, A. E., Pollard, T. J., Shen, L., Lehman, L.-w. H., Feng, M., Ghassemi, M., Moody, B., Szolovits, P., Anthony Celi, L., and Mark, R. G. (2016). MIMIC-III , a freely accessible critical care database. Scientific Data , 3(1):1--9

    2016

  28. [36]

    Li, K.-C. (1989). Honest confidence regions for nonparametric regression. The Annals of Statistics , 17(3):1001--1008

    1989

  29. [37]

    Li, S., Cai, T., and Duan, R. (2023a). Targeting underrepresented populations in precision medicine: A federated transfer learning approach. The Annals of Applied Statistics , 17(4):2970--2992

    2023

  30. [38]

    T., and Li, H

    Li, S., Cai, T. T., and Li, H. (2022). Transfer learning for high-dimensional linear regression: Prediction, estimation and minimax optimality. Journal of the Royal Statistical Society Series B: Statistical Methodology , 84(1):149--173

    2022

  31. [39]

    T., and Li, H

    Li, S., Zhang, L., Cai, T. T., and Li, H. (2023b). Estimation and inference for high-dimensional generalized linear models with knowledge transfer. Journal of the American Statistical Association , pages 1--12

    2023

  32. [40]

    C., and Zhang, Y

    Li, X., Sham, P. C., and Zhang, Y. D. (2024). A Bayesian fine-mapping model using a continuous global-local shrinkage prior with applications in prostate cancer analysis . The American Journal of Human Genetics , 111(2):213--226

    2024

  33. [41]

    Lu, Y., Gu, T., and Duan, R. (2024). Enhancing genetic risk prediction through federated semi-supervised transfer learning with inaccurate electronic health record data. Statistics in Biosciences , pages 1--22

    2024

  34. [42]

    Miglioretti, D. L. (2003). Latent transition regression for mixed outcomes. Biometrics , 59(3):710--720

    2003

  35. [43]

    Need, A. C. and Goldstein, D. B. (2009). Next generation disparities in human genomics: concerns and remedies. Trends in Genetics , 25(11):489--494

    2009

  36. [44]

    and Vehtari, A

    Piironen, J. and Vehtari, A. (2017a). On the hyperprior choice for the global shrinkage parameter in the horseshoe prior. Artificial Intelligence and Statistics , pages 905--913

    2017

  37. [45]

    and Vehtari, A

    Piironen, J. and Vehtari, A. (2017b). Sparsity information and regularization in the horseshoe and other shrinkage priors . Electronic Journal of Statistics , 11(2):5018 -- 5051

    2017

  38. [46]

    Polson, N. G. and Scott, J. G. (2011). Shrink Globally, Act Locally: Sparse Bayesian Regularization and Prediction . In Bayesian Statistics 9 . Oxford University Press

    2011

  39. [47]

    Ročkov \'a , V. (2018). Bayesian estimation of sparse signals with a continuous spike-and-slab prior . The Annals of Statistics , 46(1):401 -- 437

    2018

  40. [48]

    J., Chen, W., and Larson, N

    Schaid, D. J., Chen, W., and Larson, N. B. (2018). From genome-wide associations to candidate causal variants by statistical fine-mapping. Nature Reviews Genetics , 19(8):491--504

    2018

  41. [49]

    and Liang, F

    Song, Q. and Liang, F. (2023). Nearly optimal Bayesian shrinkage for high-dimensional regression . Science China Mathematics , 66(2):409--442

    2023

  42. [50]

    Stan modeling language users guide and reference manual, version 2.35

    Stan Development Team (2024). Stan modeling language users guide and reference manual, version 2.35

    2024

  43. [51]

    M., Xu, J., and Dunson, D

    Suder, P. M., Xu, J., and Dunson, D. B. (2023). Bayesian transfer learning. arXiv preprint arXiv:2312.13484

    2023 arXiv

  44. [52]

    M., Choi, K., and Han, P

    Taylor, J. M., Choi, K., and Han, P. (2023). Data integration: exploiting ratios of parameter estimates from a reduced external model. Biometrika , 110(1):119--134

    2023

  45. [53]

    and Feng, Y

    Tian, Y. and Feng, Y. (2023). Transfer learning under high-dimensional generalized linear models. Journal of the American Statistical Association , 118(544):2684--2697

    2023

  46. [54]

    van de Geer, S., B \"u hlmann, P., Ritov, Y., and Dezeure, R. (2014). On asymptotically optimal confidence regions and tests for high-dimensional models . The Annals of Statistics , 42(3):1166 -- 1202

    2014

  47. [55]

    van der Pas, S., Szab \'o , B., and van der Vaart, A. (2017). Uncertainty Quantification for the Horseshoe (with Discussion) . Bayesian Analysis , 12(4):1221 -- 1274

    2017

  48. [56]

    Narisetty, N., and Yang, Y

    Wu, T., N. Narisetty, N., and Yang, Y. (2023). Statistical inference via conditional Bayesian posteriors in high-dimensional linear regression . Electronic Journal of Statistics , 17(1):769--797

    2023

  49. [57]

    and Han, P

    Zhai, Y. and Han, P. (2022). Data integration with oracle use of external information from heterogeneous populations. Journal of Computational and Graphical Statistics , 31(4):1001--1012

    2022

  50. [58]

    and Zhang, S

    Zhang, C.-H. and Zhang, S. S. (2014). Confidence intervals for low dimensional parameters in high dimensional linear models. Journal of the Royal Statistical Society Series B: Statistical Methodology , 76(1):217--242

    2014

  51. [59]

    and Yin, G

    Zhang, H. and Yin, G. (2023). Unit information prior for incorporating real-world evidence into randomized controlled trials. Statistical Methods in Medical Research , 32(2):229--241

    2023

  52. [60]

    Zhang, R., Zhang, Y., Qu, A., Zhu, Z., and Shen, J. (2024). CONCERT : Covariate-elaborated robust local information transfer with conditional spike-and-slab prior. arXiv preprint arXiv:2404.03764

    2024

  53. [61]

    D., Naughton, B

    Zhang, Y. D., Naughton, B. P., Bondell, H. D., and Reich, B. J. (2022). Bayesian regression using a prior on the model fit: The R2-D2 shrinkage prior . Journal of the American Statistical Association , 117(538):862--874

    2022

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.