2026 年的期刊文章 · 第 27 页

共 1756 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2026-04-02 到 2026-03-31。

按期刊、按主题看 →

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 页

SiM Other · 发表于 2026-04-02,收录于 2026-05-01

Optimizing Medical Guidelines: Insights Using a Generalized Method of Moments Approach

Vidhura S. Tennekoon

摘要

Noncompliance with public health guidelines is a widely discussed public health issue, but it can also reflect rational, utility‐maximizing behavior. Using 41.1 million birth records from 2011–2021, we explore three possible explanations: asymmetric information, divergent objectives between experts and the public, and limitations in the guidelines themselves. We model the expert committee's objective function as minimizing an aggregate health risk score conditional on prepregnancy BMI and estimate its underlying risk weights by developing and using a Generalized Method of Moments (GMM) estimator. We then compare these with the risk preferences revealed by birthing people. Our findings show that experts implicitly assign equal weight to the risks of low and high birth weight, whereas birthing people place less weight on the risk of high birth weight. We then demonstrate that the current guidelines are suboptimal—even under the criteria they were designed to meet. We propose a new, individualized guideline structure that adjusts continuously with prepregnancy BMI. These improved guidelines not only reduce aggregate health risks but also better reflect birthing people's preferences—offering a path to significantly higher compliance and more effective public health policy. The paper illustrates the broader applicability of the GMM framework for recovering latent preferences in structural models of health behavior. JEL Classification: I12, I15, J18

JASA Optimization & Learning Theory · 发表于 2026-04-02

Online Statistical Inference for Contextual Bandits via Stochastic Gradient Descent

Xiangyu Chang, Xi Chen, Zehua Lai, He Li, Zhihong Liu, Yichen Zhang

摘要

With the fast development of big data, learning the optimal decision rule by recursively updating it and making online decisions has been easier than before. We study the online statistical inference of model parameters in a contextual bandit framework of sequential decision-making. We propose a general framework for an online and adaptive data collection environment that can update decision rules via weighted stochastic gradient descent. We allow different weighting schemes of the stochastic gradient and establish the asymptotic normality of the parameter estimator. Our proposed estimator significantly improves the asymptotic efficiency over the previous averaged SGD approach via inverse probability weights. We also conduct an optimality analysis on the weights in a linear regression setting. We provide a Bahadur representation of the proposed estimator and show that the remainder term in the Bahadur representation entails a slower convergence rate compared to classical SGD due to the adaptive data collection. *(abstract via OpenAlex)*

JASA Causal Inference · 发表于 2026-04-02

On the Identifying Power of Generalized Monotonicity for Average Treatment Effects

Yuehao Bai, Shunzhuang Huang, Sarah Moon, Azeem M. Shaikh, Edward J. Vytlacil

摘要

In the context of a binary outcome, treatment, and instrument, Balke and Pearl establish that the monotonicity condition of Imbens and Angrist has no identifying power beyond instrument exogeneity for average potential outcomes and average treatment effects in the sense that adding it to instrument exogeneity does not decrease the identified sets for those parameters whenever those restrictions are consistent with the distribution of the observable data. This article shows that this phenomenon holds in a broader setting with a multi-valued outcome, treatment, and instrument, under an extension of the monotonicity condition that we refer to as generalized monotonicity. We further show that this phenomenon holds for any restriction on treatment response that is stronger than generalized monotonicity provided that these stronger restrictions do not restrict potential outcomes. Importantly, many models of potential treatments previously considered in the literature imply generalized monotonicity. We show through a series of examples that restrictions on potential treatments can provide identifying power beyond instrument exogeneity for average potential outcomes and average treatment effects when the restrictions imply that the generalized monotonicity condition is violated. In this way, our results shed light on the types of restrictions required for help in identifying average potential outcomes and average treatment effects. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*

JRSSB Survival Analysis Machine Learning Networks & Graphs · 发表于 2026-04-02

Nonparametric inference for censored data using deep neural networks

Wen Su, Qiang Wu, Kin-Yat Liu, Guosheng Yin, Jian Huang, Xingqiu Zhao

摘要

We propose a novel deep learning approach to nonparametric statistical inference for the conditional hazard function of survival time with right-censored data. We use a deep neural network (DNN) to approximate the logarithm of a conditional hazard function given covariates and obtain a DNN likelihood-based estimator of the conditional hazard function. Such an estimation approach enhances model flexibility and hence relaxes structural and functional assumptions on conditional hazard or survival functions. We establish the nonasymptotic error bound and functional asymptotic normality of the proposed estimator. Subsequently, we develop new one-sample tests for goodness-of-fit evaluation and two-sample tests for treatment comparison. Notably, we design a new test specifically tailored for testing nonparametric Cox models. The consistency of these tests is established by analyzing the power functions. Both simulation studies and real application analysis show superior performances of the proposed estimators and tests in comparison with existing methods.

SiM Bayesian Methods Nonparametric & Semiparametric Spatial & Time Series · 发表于 2026-04-02,收录于 2026-05-01

Nonparametric Bayesian Meta‐Analysis Model With Change Point Detection: A Case Study on Time‐Varying Temperature–Mortality Associations

Daewon Yang, Taeryon Choi, Jinsu Park, Hohyun Jung, Yoonhee Kim, Masahiro Hashizume, Yeonseung Chung

摘要

Numerous studies have established a U‐ or J‐shaped association between ambient temperature and human mortality by analyzing time‐series data collected from multiple locations. However, this association has changed over time owing to climate change and adaptive behaviors, such as the increased use of air conditioning. Temporal changes have been examined using a conventional two‐stage modeling framework, which typically assumes a linear change over time. Our preliminary analysis of Japanese data suggests that these temporal changes may not be linear or gradual, challenging the common assumption adopted in the conventional two‐stage modeling approach. Moreover, the assumption of normality in the existing second‐stage mixed‐effects models can make the estimates highly sensitive to outliers. To address these limitations, we propose a two‐stage modeling framework with a novel second stage model: a nonparametric Bayesian meta‐analysis model with change‐point detection. In the first stage, we divide the full study period into non‐overlapping one‐year sub‐periods to capture fine‐scale temporal variation and estimate temperature–mortality associations for each location and sub‐period using distributed lag nonlinear models. In the second stage, we pool the associations across locations and sub‐periods simultaneously applying our novel nonparametric Bayesian meta‐analytic model, which is formulated based on the Probit Stick‐Breaking Process to facilitate flexible segmentation of time regimes. The model also incorporates a mixture of Gaussian and t‐distributed errors to account for potential outliers and ensure robust estimation. The proposed modeling framework is validated through simulation studies and illustrated through an application to the study for time‐varying temperature–mortality association in Japan.

JRSSB Other · 发表于 2026-04-02

Model privacy: a unified framework for understanding model stealing attacks and defences

Ganghua Wang, Yuhong Yang, Jie Ding

摘要

The use of machine learning (ML) has become increasingly prevalent in various domains, highlighting the importance of understanding and ensuring its safety. One pressing concern is the vulnerability of ML applications to model stealing attacks. These attacks involve adversaries attempting to recover a learned model through limited query-response interactions, such as those found in cloud-based services or on-chip artificial intelligence interfaces. While existing literature proposes various attack and defence strategies, these often lack a theoretical foundation and standardized evaluation criteria. In response, this work presents a framework called ‘Model Privacy’, providing a foundation for comprehensively analyzing model stealing attacks and defences. We establish a rigorous formulation for the threat model and objectives, propose methods to quantify the goodness of attack and defence strategies, and analyse the fundamental tradeoffs between utility and privacy in ML models. Our developed theory offers valuable insights into enhancing the security of ML models, especially highlighting the importance of the query-dependent structure of perturbations for effective defences. We demonstrate the application of model privacy from the defender’s perspective through various learning scenarios. Extensive experiments corroborate the insights and the effectiveness of defence mechanisms developed under the proposed framework.

SMMR Other · 发表于 2026-04-02

Linearized maximum rank correlation estimation of doubly truncated data

Peijie Wang, Qihao Wang, Jianguo Sun

摘要

Truncated data frequently arise in many areas such as economics, astronomical studies, and survival analysis, and the existence of truncation makes statistical inference more difficult due to the incomplete information. In this paper, we propose a linearized maximum rank correlation estimation of doubly truncated data under a single-index model. Unlike the existing methods, the proposed estimation has a closed-form expression and does not need knowledge of the unknown link function or the error distribution, which makes it more appealing in theory and computation. The proposed estimators are shown to be consistent and asymptotically normal, and an extensive simulation study is conducted and indicates that the proposed method works well in various situations. The method is further demonstrated by applying it to an AIDS study.

SiM Bayesian Methods Survival Analysis High-Dimensional · 发表于 2026-04-02,收录于 2026-05-01

Flexible yet Sparse Bayesian Survival Models With Time‐Varying Coefficients and Unobserved Heterogeneity

Peter Knaus, Daniel Winkler, Sebastian F. Schoppmann, Gerd Jomrich

摘要

Survival analysis is an important area of medical research, yet existing models often struggle to balance simplicity with flexibility. Simple models require minimal adjustments but come with strong assumptions, while more flexible models require significant input and tuning from researchers. We present a survival model using a Bayesian hierarchical shrinkage method that automatically determines whether each covariate should be treated as static, time‐varying, or excluded altogether. This approach strikes a balance between simplicity and flexibility, minimizes the need for tuning, and naturally quantifies uncertainty. The method is supported by an efficient Markov chain Monte Carlo sampler, implemented in the R package shrinkDSM. Comprehensive simulation studies and an application to a clinical dataset involving patients with adenocarcinoma of the gastroesophageal junction showcase the advantages of our approach compared to existing models.

JASA Spatial & Time Series · 发表于 2026-04-02

Fast and Flexible Emulation of Spatial Extremes Processes via Variational Autoencoders

Likun Zhang, Xiaoyu Ma, Christopher K. Wikle, Raphaël Huser

摘要

Many real-world processes have complex tail dependence structures that cannot be adequately characterized using classical Gaussian processes. Alternatively, models motivated by extreme-value theory exhibit appealing extremal dependence properties but are often exceedingly prohibitive to fit and simulate from in high dimensions using classical methods. In this paper, we extend the boundaries on computation and modeling of high-dimensional spatial extremes by integrating a new flexible and nonstationary spatial extremes model into the encoding-decoding structure of a variational autoencoder, called the extremes variational autoencoder (XVAE). Our proposed model is not only able to capture asymptotic dependence and independence, but also spatio-temporal heterogeneity in its tail behavior. The proposed XVAE is designed to realistically emulate spatial observations and produce outputs that have similar statistical properties as the inputs, especially in the tail. By exploiting our neural-network-based approach, we also provide a novel way of making fast inference with complex extreme-value processes. Through extensive simulation studies, we show that the XVAE is substantially more time-efficient than traditional Bayesian inference methods and outperforms other competing modeling approaches. Lastly, we analyze a high-resolution satellite-derived dataset of sea surface temperature in the Red Sea, which includes 30 years of daily measurements at 16 703 grid cells. We demonstrate how the XVAE can be used to identify regions susceptible to marine heatwaves and how these have evolved during the study period. *(abstract via OpenAlex)*

SiM Causal Inference · 发表于 2026-04-02,收录于 2026-04-24

Examining the Efficacy of Coarsened Exact Matching as an Alternative to Propensity Score Matching

Fei Wan

摘要

Coarsened exact matching (CEM) is often promoted as a superior alternative to propensity score matching (PSM) for addressing imbalance, model dependence, bias, and efficiency. However, this recommendation remains uncertain. First, CEM is commonly mischaracterized as exact matching, despite relying on coarsened rather than original variables. This inexactness in matching introduces residual confounding, which necessitates accurate modeling of the outcome‐confounder relationship post‐matching to mitigate bias, thereby increasing vulnerability to model misspecification. Second, prior studies overlook that any imbalance between treated and untreated subjects matched on the same propensity score is attributable to random variation. Thus, claims that CEM outperforms PSM in reducing imbalance are unfounded, particularly when using metrics like Mahalanobis distance, which do not account for chance imbalance in PSM. Our simulations show that PSM reduces imbalance more effectively than CEM when evaluated with multivariate standardized mean differences (SMD), and unadjusted analyses indicate greater bias with CEM. While adjusted analyses in both CEM with autocoarsening and PSM may perform similarly when matching on a few variables, CEM suffers from the curse of dimensionality as the number of factors increases, resulting in substantial data loss and unstable estimates. Increasing the level of coarsening may mitigate data loss but exacerbates residual confounding and model dependence. In contrast, both analytical results and simulations demonstrate that PSM is more robust to model misspecification and thus less model‐dependent. Therefore, CEM is not a viable alternative to PSM when matching on a large number of covariates.

SiM Survival Analysis Bayesian Methods · 发表于 2026-04-02,收录于 2026-05-01

Bayesian Random‐Effects Meta‐Analysis of Aggregate Data on Clinical Events

Christian Röver, Qiong Wu, Anja Loos, Tim Friede

摘要

(verbatim, in full — via PMC, cross-confirmed against the arXiv preprint arXiv:2504.12214): "To investigate intervention effects on rare events, meta‐analysis techniques are commonly applied in order to assess the accumulated evidence. When it comes to adverse effects in clinical trials, these are often most adequately handled using survival methods. A common‐effect model that is able to process data in commonly quoted formats in terms of hazard ratios has previously been proposed for this purpose. In order to accommodate potential heterogeneity between studies, we have extended the model by Holzhauer to a random‐effects approach. The Bayesian model is described in detail, and applications to realistic data sets are discussed along with sensitivity analyses and Monte Carlo simulations to support the conclusions.

SiM Causal Inference Bayesian Methods · 发表于 2026-04-02,收录于 2026-05-01

Bayesian Hierarchical Models With Calibrated Mixtures of g‐priors for Assessing Treatment Effect Moderation in Meta‐Analysis

Qiao Wang, Hwanhee Hong

摘要

(verbatim from arXiv:2410.24194 — onlinelibrary.wiley.com returned 403, so this is the preprint abstract; exact online date not confirmed)*: "Assessing treatment effect moderation is critical in biomedical research and many other fields, as it guides personalized intervention strategies to improve participant's outcomes. Individual participant-level data meta-analysis (IPD-MA) offers a robust framework for such assessments by leveraging data from multiple trials. However, its performance is often compromised by challenges such as high between-trial variability. Traditional Bayesian shrinkage methods have gained popularity, but are less suitable in this context, as their priors do not discern heterogeneous studies. In this paper, we propose the calibrated mixtures of g-priors methods in IPD-MA to enhance efficiency and reduce risk in the estimation of moderation effects. Our approach incorporates a trial-level sample size tuning function, and a moderator-level shrinkage parameter in the prior, offering a flexible spectrum of shrinkage levels that enables practitioners to evaluate moderator importance, from conservative to optimistic perspectives. Compared with existing Bayesian shrinkage methods, our extensive simulation studies demonstrate that the calibrated mixtures of g-priors exhibit superior performances in terms of efficiency and risk metrics, particularly under high between-trial variability, high model sparsity, weak moderation effects and correlated design matrices. We further illustrate their application in assessing effect moderators of two active treatments for major depressive disorder, using IPD from four randomized controlled trials.

AoS Other · 发表于 2026-04-01

Versatile differentially private learning for general loss functions

Qilong Lu, Song Xi Chen, Yumou Qiu

摘要

This paper aims to provide a versatile privacy-preserving release mechanism along with a unified approach for subsequent parameter estimation and statistical inference. We propose a privacy mechanism based on zero-inflated symmetric multivariate Laplace (ZIL) noise, which requires no prior specification of subsequent analysis tasks, allows for general loss functions under minimal conditions, imposes no limit on the number of analyses, and is adaptable to increasing data volume in online scenarios. We derive the trade-off function for the proposed ZIL mechanism, which characterizes its privacy protection level. Furthermore, to formalize the local differential privacy (LDP) property of the ZIL mechanism, we extend the classical ε-LDP to a more general f-LDP framework. To address scenarios where only individual attribute values require protection, we propose attribute-level differential privacy (ADP) and its local version. Within the M-estimation framework, we introduce a novel doubly random (DR) corrected loss for the ZIL mechanism, which yields consistent and asymptotically normal M-estimates under differential privacy constraints. The proposed approach is computationally efficient and does not require numerical integration or differentiation for noisy data. It applies to a broad class of loss functions, including nonsmooth ones. Two alternative estimators for smooth loss are also proposed with asymptotic properties. The cost of privacy in terms of estimation efficiency for these three estimators is evaluated both theoretically and numerically.

AoS High-Dimensional · 发表于 2026-04-01

Trace test for high-dimensional cointegration

Alexei Onatski, Chen Wang

摘要

This paper studies Johansen’s (J. Econom. Dynam. Control 12 (1988) 231–254) trace test for cointegration in high-dimensional data. We show that when both cross-sectional and temporal dimension of the data go to infinity proportionally, the shifted and scaled modified trace statistic converges to a Gaussian random variable. We give explicit formulae for the shift and scale parameters as well as for the mean and variance of the Gaussian limit. Monte Carlo analysis shows excellent size properties of the asymptotic test, which is an improvement over the Bartlett-corrected versions of the original trace test, especially for relatively large ratios of the dimensionality to the sample size. The Monte Carlo also reveals a nonmonotonicity of the power of the test. We comment on the source of such a nonmonotonicity.

AoS High-Dimensional · 发表于 2026-04-01

The distributionally robust prediction error of the LASSO and related estimators

José Luis Montiel Olea, Cynthia Rush, Amilcar Velez, Johannes Wiesel

摘要

We study the classical problem of predicting an outcome variable, Y, using a linear combination of a d-dimensional covariate vector, X. We are interested in linear predictors whose coefficients solve infβ∈Rd(EPn[|Y−X⊤β|r])1/r+δρ(β), where δ>0 is a regularization parameter, ρ:Rd→R+ is a convex penalty function, Pn is the empirical distribution of the data and r≥1. Our main contribution is a new bound on the out-of-distribution prediction error of such estimators. The new bound is obtained by combining three new sets of results. First, we provide conditions under which linear predictors based on these estimators solve a distributionally robust optimization problem: they minimize the worst-case prediction error over distributions that are close to each other in a type of max-sliced Wasserstein metric. Second, we provide a detailed finite-sample and asymptotic analysis of the statistical properties of the balls of distributions over which the worst-case prediction error is analyzed. Third, we present an oracle recommendation for the choice of regularization parameter, δ, that guarantees good out-of-distribution prediction error.

AoS High-Dimensional · 发表于 2026-04-01

Spectrum-aware debiasing: A modern inference framework with applications to principal components regression

Yufan Li, Pragya Sur

摘要

Debiasing is a fundamental concept in high-dimensional statistics. While degrees-of-freedom adjustment is the state-of-the-art debiasing technique in high-dimensional linear regression, it largely remains limited to independent, identically distributed samples and sub-Gaussian covariates. These limitations hinder its wider practical use. In this paper we break this barrier and introduce Spectrum-Aware Debiasing—a novel inference method that applies to challenging high-dimensional regression problems with structured row-column dependencies, heavy tails, asymmetric properties, and latent low-rank structures. Our method achieves debiasing through a rescaled gradient descent step, where the rescaling factor is derived from the spectral properties of the sample covariance matrix. This spectrum-based approach enables accurate debiasing in much broader contexts. We study the common modern regime where the number of features and samples scale proportionally. We establish asymptotic normality of our proposed estimator (suitably centered and scaled) under various convergence notions when the covariates are right-rotationally invariant. We further prove a spectral universality result, extending our guarantees to a much broader class of covariate distributions. Furthermore, we devise a consistent estimator for the asymptotic variance. Our work has two notable by-products: First, Spectrum-Aware Debiasing rectifies the bias in principal components regression (PCR), providing the first debiased PCR estimator in high dimensions. Second, we introduce a principled test for checking the presence of alignment between the signal and the eigenvectors of the sample covariance matrix. This test is independently valuable for statistical methods developed using approximate message passing, leave-one-out, random matrix theory, or convex Gaussian min-max theorems. We demonstrate the utility of our method through diverse simulated and real data experiments.

JASA Longitudinal Data Survival Analysis Nonparametric & Semiparametric · 发表于 2026-04-01,收录于 2026-09-10

Semiparametric Joint Modeling for Survival Analysis With Longitudinal Covariates

NOT RETRIEVED — tandfonline byline was inaccessible, and no matching preprint could be located to cross-check authorship.

摘要

Characterizing the association between survival time and the dynamic patterns of a longitudinal covariate trajectory is of particular interest in many studies. Classical time-dependent survival models focus mainly on the link between the concurrent covariate value and the instantaneous hazard function. Consequently, the conditional survival function is often not properly defined on the whole time range, which causes difficulty in model estimation and interpretation. In this article, we propose a novel semiparametric joint modeling approach, in which the observed longitudinal trajectory is modeled as a random realization of a latent functional pattern. We assume each latent pattern uniquely indexes a global survival function via a log-linear functional regression model. Because the observational time interval of the longitudinal data depends on the survival time, we propose to jointly model the longitudinal and survival data. By using the latent pattern as an infinite-dimensional shared parameter, our approach extends the classical parametric joint modeling method to a semiparametric setting. We show that the proposed estimator achieves the semiparametric efficiency bound. Simulation studies and a real data application demonstrate the advantageous finite sample performances of our new approach. *(abstract via PubMed)*

AoS Nonparametric & Semiparametric Testing & Uncertainty · 发表于 2026-04-01

Scalable inference for nonparametric stochastic approximation in reproducing kernel Hilbert spaces

Meimei Liu, Zuofeng Shang, Yun Yang

摘要

Stochastic approximation (SA) is a powerful and scalable computational method for iteratively estimating the solution of optimization problems in the presence of randomness, particularly well suited for large-scale and streaming data settings. In this work we propose a theoretical framework for stochastic approximation (SA) applied to nonparametric least squares in reproducing kernel Hilbert spaces (RKHS), enabling online statistical inference in nonparametric regression models. We achieve this by constructing asymptotically valid pointwise (and simultaneous) confidence intervals (bands) for local (and global) inference of the nonlinear regression function, via employing an online multiplier bootstrap approach to a functional stochastic gradient descent (SGD) algorithm in the RKHS. Our main theoretical contributions consist of a unified framework for characterizing the nonasymptotic behavior of the functional SGD estimator and demonstrating the consistency of the multiplier bootstrap method. The proof techniques involve the development of a higher-order expansion of the functional SGD estimator under the supremum norm metric and the Gaussian approximation of suprema of weighted and nonidentically distributed empirical processes. Our theory specifically reveals an interesting relationship between the tuning of step sizes in SGD for estimation and the accuracy of uncertainty quantification.

AoS Other · 发表于 2026-04-01

Reviving pseudo-inverses: Asymptotic properties of large dimensional Moore–Penrose and ridge-type inverses with applications

Taras Bodnar, Nestor Parolya

摘要

In this paper, we derive high-dimensional asymptotic properties of the Moore–Penrose inverse and, as a byproduct, of various ridge-type inverses of the sample covariance matrix. In particular, the analytical expressions of the asymptotic behavior of the weighted sample trace moments of generalized inverse matrices are deduced in terms of the partial exponential Bell polynomials, which can be easily computed in practice. The existent results for pseudo-inverses are extended in several directions: (i) First, the population covariance matrix is not assumed to be a multiple of the identity matrix; (ii) Second, the assumption of normality is not used in the derivation; (iii) Third, the asymptotic results are derived under the high-dimensional asymptotic regime. Our findings provide universal methodology for construction of fully data-driven improved shrinkage estimators of the precision matrix, optimal portfolio weights and beyond. It is found that the Moore–Penrose inverse acts asymptotically as a certain regularizer of the true covariance matrix and it seems that its proper transformation (shrinkage) performs similar to or even outperforms the existing benchmarks in many applications, while keeping the computational time as minimal as possible.

AoS Other · 发表于 2026-04-01

Rank tests for PCA under weak identifiability

Davy Paindaveine, Laura Peralvo Maroto, Thomas Verdebout

摘要

In a triangular array framework where n observations are randomly sampled from a p-dimensional elliptical distribution with shape matrix Vn, we consider the problem of testing the null hypothesis H0:θ=θ0 against the alternative hypothesis H1:θ≠θ0, where θ is the (fixed) leading unit eigenvector of Vn and θ0 is a given unit p-vector. The dependence of the shape matrix on the sample size allows us to consider challenging asymptotic scenarios in which the parameter of interest θ is unidentified in the limit, because the ratio between both leading eigenvalues of Vn converges to one. We carefully study the corresponding limiting experiments under such weak identifiability, and we show that these may be LAN or non-LAN. While earlier work in the framework was strictly limited to Gaussian distributions, where the study of local log-likelihood ratios could simply rely on explicit expressions, our asymptotic investigation allows for essentially arbitrary elliptical distributions. This requires original results on quadratic mean differentiable families for triangular arrays of observations, which are likely to be of interest in other models, too. Even in non-LAN experiments, our results enable us to investigate, through Le Cam’s first and third lemmas, the asymptotic null and nonnull properties of multivariate rank tests. These nonparametric tests are shown to exhibit an excellent behavior under weak identifiability: not only do they maintain the target nominal size irrespective of the amount of weak identifiability, but they also keep their outstanding uniform efficiency properties under such nonstandard scenarios. In particular, Gaussian-score rank tests, under arbitrarily weak identifiability, still uniformly dominate their parametric pseudo-Gaussian competitor in terms of asymptotic relative efficiencies. Our theoretical results, which are the first ones to study rank tests in the triangular array framework allowing for weak identifiability, are supported by several Monte Carlo exercises.

AoS Clinical Trials · 发表于 2026-04-01

Precise asymptotics of bagging regularized M-estimators

Takuya Koriyama, Pratik Patil, Jin-Hong Du, Kai Tan, Pierre C. Bellec

摘要

We characterize the squared prediction risk of ensemble estimators obtained through subagging (subsample bootstrap aggregating) regularized M-estimators and construct a consistent estimator for the risk. Specifically, we consider a heterogeneous collection of M≥1 regularized M-estimators, each trained with (possibly different) subsample sizes, convex differentiable losses, and convex regularizers. We operate under the proportional asymptotics regime, where the sample size n, feature size p, and subsample sizes km for m∈[M] all diverge with fixed limiting ratios n/p and km/n. Key to our analysis is a new result on the joint asymptotic behavior of correlations between the estimator and residual errors on overlapping subsamples, governed through a (provably) contractive nonlinear system of equations. Of independent interest we also establish convergence of trace functionals related to degrees of freedom in the nonensemble setting (with M=1) along the way, extending previously known cases for squared loss with ridge and lasso regularizers. When specialized to homogeneous ensembles trained with a common loss, regularizer, and subsample size, the risk characterization sheds some light on the (implicitly) induced regularization effect due to the ensemble and subsample sizes (M,k). For any ensemble size M, optimally tuning subsample size yields samplewise monotonic risk. For the full-ensemble estimator (when M→∞), the optimal subsample size k⋆ tends to be in the overparameterized regime (k⋆≤min{n,p}), when explicit regularization is vanishing. Finally, joint optimization of subsample size, ensemble size, and regularization can significantly outperform regularizer optimization alone on the full data (without any subagging).

AoS Causal Inference Latent Variables & Mixtures · 发表于 2026-04-01

Parameter identification in linear non-Gaussian causal models under general confounding

Daniele Tramontano, Mathias Drton, Jalal Etesami

摘要

Linear non-Gaussian causal models postulate that each random variable is a linear function of parent variables and non-Gaussian exogenous error terms. We study identification of the linear coefficients when such models contain latent variables. Our focus is on the commonly studied acyclic setting, where each model corresponds to a directed acyclic graph (DAG). For this case, prior literature has demonstrated that connections to overcomplete independent component analysis yield effective criteria to decide parameter identifiability in latent variable models. However, this connection is based on the assumption that the observed variables linearly depend on the latent variables. Departing from this assumption, we treat models that allow for arbitrary nonlinear latent confounding. Our main result is a graphical criterion that is necessary and sufficient for deciding the generic identifiability of direct causal effects. Moreover, we provide an algorithmic implementation of the criterion with a run time that is polynomial in the number of observed variables. Finally, we report on estimation heuristics based on the identification result and explore a generalization to models with feedback loops.

AoS Spatial & Time Series · 发表于 2026-04-01

PCA for point processes

Franck Picard, Vincent Rivoirard, Angelina Roche, Victor M. Panaretos

摘要

We introduce a novel statistical framework for the analysis of replicated point processes that allows for the study of point pattern variability at a population level. By treating point process realizations as random measures, we adopt a functional analysis perspective and propose a form of functional Principal Component Analysis (fPCA) for point processes. The originality of our method is to base our analysis on the cumulative mass functions of the random measures, which gives us a direct and interpretable analysis. Key theoretical contributions include establishing a Karhunen–Loève expansion for the random measures and a Mercer theorem for covariance measures. We establish convergence in a strong sense, and introduce the concept of principal measures, which can be seen as latent processes governing the dynamics of the observed point patterns. We propose an easy-to-implement estimation strategy of eigenelements for which parametric rates are achieved. We fully characterize the solutions of our approach to Poisson and Hawkes processes and validate our methodology via simulations and diverse applications in seismology, single-cell biology and neurosiences, demonstrating its versatility and effectiveness. Our method is implemented in the pppca R-package.

AoS Other · 发表于 2026-04-01

Optimal eigenvalue shrinkage in the semicircle limit

David L. Donoho, Michael J. Feldman

摘要

In response to the increasing dimensionality of modern datasets, recent theoretical studies of covariance estimation frequently adopt the proportional-growth asymptotic framework in which the sample size n and dimension p are comparable, with n,p→∞ and γn:=p/n→γ>0. However, in many datasets—perhaps most—the sample size and dimension are highly imbalanced. To address this, we consider the disproportional-growth asymptotic framework, where n,p→∞ and γn→0 or γn→∞. These regimes give rise to novel behavior distinct from those in the proportional-growth and classical fixed-p settings. We work under the spiked covariance model in which the theoretical covariance matrix is a low-rank perturbation of the identity. For each of 15 loss functions, we derive closed-form optimal shrinkage and thresholding rules; for several losses, optimality takes the particularly strong form of unique asymptotic admissibility. These optimal procedures involve substantial eigenvalue shrinkage and yield significant improvements over the standard empirical covariance estimator. Practitioners may ask whether their data is better modeled under the proportional or disproportional frameworks and which of the corresponding procedures to apply. Fortunately, it is possible to remain framework-agnostic: a single, unified set of shrinkage rules—depending only on the aspect ratio γn of the given data—achieves asymptotic optimality in both regimes. At the heart of these phenomena is the spiked Wigner model in which a low-rank matrix is perturbed by symmetric noise. Under both the (scaled) spiked covariance model as γn→0 and the spiked Wigner model, the empirical spectral distributions converge to the semicircle law. Exploiting this connection, we derive optimal spiked-Wigner shrinkage rules, which are of independent and fundamental interest.

AoS Other · 发表于 2026-04-01

Object detection under the linear subspace model with application to cryo-EM images

Keren Mor Waknin, Amitay Eldar, Samuel Davenport, Tamir Bendory, Armin Schwartzman, Yoel Shkolnisky

摘要

Detecting multiple unknown objects in noisy data is a key problem in many scientific fields, such as electron microscopy imaging. A common model for the unknown objects is the linear subspace model, which assumes that the objects can be expanded in some known basis (such as the Fourier basis). In this paper, we develop an object detection algorithm that under the linear subspace model is asymptotically guaranteed to detect all objects, while controlling the familywise error rate or the false discovery rate. Numerical simulations show that the algorithm also controls the error rate with high power in the nonasymptotic regime, even in highly challenging regimes. We apply the proposed algorithm to an experimental electron microscopy data set, and show that it outperforms existing standard software.

AoS Testing & Uncertainty Econometrics & Economics · 发表于 2026-04-01

Large-scale multiple testing: Fundamental limits of false discovery rate control and compound oracle

Yutong Nie, Yihong Wu

摘要

The false discovery rate (FDR) and the false nondiscovery rate (FNR), defined as the expected false discovery proportion (FDP) and the false nondiscovery proportion (FNP), are the most popular benchmarks for multiple testing. Despite the theoretical and algorithmic advances in recent years, the optimal trade-off between the FDR and the FNR has been largely unknown, except for certain restricted classes of decision rules, for example, separable rules, or for other performance metrics, for example, the marginal FDR and the marginal FNR (mFDR and mFNR). In this paper we determine the asymptotically optimal FDR-FNR trade-off under the two-group random mixture model when the number of hypotheses tends to infinity. Distinct from the optimal mFDR-mFNR trade-off, which is achieved by separable decision rules, the optimal FDR-FNR trade-off requires compound rules, even in the large-sample limit and for models as simple as the Gaussian location model. This suboptimality of separable rules also holds for other objectives, such as maximizing the expected number of true discoveries. Finally, to address the limitation of the FDR, which only controls the expectation but not the fluctuation of the FDP, we also determine the optimal tradeoff when the FDP is controlled with high probability and show it coincides with that of the mFDR and the mFNR. Extensions to models with a fixed nonnull proportion are also obtained.

AoS Networks & Graphs High-Dimensional · 发表于 2026-04-01

Inferring the dependence graph density of binary graphical models in high dimension

Julien Chevallier, Eva Löcherbach, Guilherme Ost

摘要

We consider a system of binary interacting chains describing the dynamics of a group of N components that, at each time unit, either send some signal to the others or remain silent otherwise. The interactions among the chains are encoded by a directed Erdös–Rényi random graph with unknown parameter p∈(0,1). Moreover, the system is structured within two populations (excitatory chains versus inhibitory ones), which are coupled via a mean field interaction on the underlying Erdös–Rényi graph. In this paper, we address the question of inferring the connectivity parameter p based only on the observation of the interacting chains over T time units. In our main result, we show that the connectivity parameter p can be estimated with rate N−1/2+N1/2/T+(log(T)/T)1/2 through an easy-to-compute estimator. Our analysis relies on a precise study of the spatiotemporal decay of correlations of the interacting chains. This is done through the study of coalescing random walks defining a backward regeneration representation of the system. Interestingly, we also show that this backward regeneration representation allows us to perfectly sample the system of interacting chains (conditionally on each realization of the underlying Erdös–Rényi graph) from its stationary distribution. These probabilistic results have an interest in its own.

AoS Other · 发表于 2026-04-01

Generalized multilinear models for sufficient dimension reduction on tensor-valued predictors

Daniel Kapla, Efstathia Bura

摘要

We consider supervised learning problems with tensor-valued input. We derive multilinear sufficient reductions for the regression or classification problem by modeling the conditional distribution of the predictors given the response as a member of the quadratic exponential family. We develop estimation procedures of sufficient reductions for both continuous and binary tensor-valued predictors. We prove the consistency and asymptotic normality of the estimated sufficient reduction using manifold theory. For multilinear normal predictors, the estimation algorithm is highly computationally efficient and is also applicable to situations where the dimension of the reduction exceeds the sample size. Our method outperforms competing techniques in both simulated settings and real-world datasets involving continuous and binary tensor-valued predictors.

AoS High-Dimensional · 发表于 2026-04-01

Generalized linear spectral statistics of high-dimensional sample covariance matrices and its applications

Yanlin Hu, Qing Yang, Xiao Han

摘要

In this paper, we introduce the Generalized Linear Spectral Statistics (GLSS) of a high-dimensional sample covariance matrix Sn, denoted as trf(Sn)Bn, which effectively captures distinct spectral properties of Sn by incorporating an ancillary matrix Bn and a test function f. The joint asymptotic normality of GLSS associated with different test functions is established under mild assumptions on Bn and the underlying distribution, when the dimension n and sample size N are comparable. The convergence rate of GLSS is determined by N/rank(B n). Subsequently, we propose a novel functional projection approach based on GLSS for hypothesis testing on eigenspaces of “population-spiked” covariance matrices, showcasing a universality phenomenon in the magnitude of the spikes. The theoretical accuracy of our results established for GLSS and the advantages of the newly suggested testing procedure are demonstrated through various numerical studies.

AoS High-Dimensional Testing & Uncertainty · 发表于 2026-04-01

Finite- and large sample inference for model and coefficients in high-dimensional linear regression with repro samples

Peng Wang, Min-ge Xie, Linjun Zhang

摘要

In this paper, we present a novel and effective inference approach to conduct both finite and large sample inference for high-dimensional linear regression models. This approach is developed under the so-called repro samples framework, in which we conduct statistical inference by creating and studying the behavior of artificial samples that are obtained by mimicking the sampling mechanism of the data. We construct confidence sets for (a) the true model corresponding to the nonzero coefficients, (b) a single or any collection of regression coefficients and (c) both the model and regression coefficients jointly. To facilitate the constructions of these confidence sets and overcome computational difficulties of searching all possible models, we use an innovative Fisher inversion technique to construct a model candidate set that includes the true sparse model with probability close to 1 for models with both Gaussian and non-Gaussian errors. The proposed approach fills in two major gaps in the high-dimensional regression literature: (1) lack of effective approaches to addressing model selection uncertainty and providing valid inference for the underlying true model; (2) lack of effective inference approaches to guarantee finite sample performance. We provide both finite sample and asymptotic results to theoretically guarantee the performance of the proposed methods. In addition, our numerical results demonstrate that the proposed methods are valid and achieve better coverage with smaller confidence sets than the current state-of-the-art approaches, such as debiasing and bootstrap approaches.

SiM Diagnostics & Prediction Nonparametric & Semiparametric Testing & Uncertainty · 发表于 2026-04-01,收录于 2026-05-01

Evaluation of AI‐Based Medical Device Concerning Localization Information Using Nonparametric Inference for the Alternative Free‐Response ROC Curve

Kaiyuan Liu, Xiao‐Hua Zhou

摘要

The alternative free‐response receiver operating characteristic (AFROC) curve is a popular method for evaluating the performance of diagnostic tests concerning detecting and locating abnormal lesions. However, the existing inferences for the AFROC curve rely on the assumptions of the independence of the observations within the same subject and certain parametric models, which are hard to test and may not be true in practice. In this article, we propose nonparametric inferences for the AFROC curve. Under reasonable assumptions, we derive the asymptotic properties for the empirical AFROC curve and use them to make inferences about the AFROC curve and its related indices. We propose a new bootstrap method to construct the confidence intervals of the indices related to the AFROC curve and the confidence band for the AFROC curve. Simulations show that our method significantly outperforms the existing parametric approach when its assumptions are violated. We also provide a real‐world example, that is, a diagnostic test investigating AI‐assisted pulmonary nodule diagnosis, to illustrate the practical applicability of our method.

AoS Networks & Graphs Spatial & Time Series Latent Variables & Mixtures · 发表于 2026-04-01

Estimation of grouped time-varying network vector autoregressive models

Degui Li, Bin Peng, Songqiao Tang, Weibiao Wu

摘要

This paper introduces a flexible time-varying network vector autoregressive model framework for large-scale time series. A latent group structure is imposed on the heterogeneous and node-specific time-varying momentum and network spillover effects so that the number of unknown time-varying coefficients to be estimated can be reduced considerably. A classic agglomerative clustering algorithm with nonparametrically estimated distance matrix is combined with a ratio criterion to consistently estimate the latent group number and membership. A postgrouping local linear smoothing method is proposed to estimate the group-specific time-varying momentum and network effects, substantially improving the convergence rates of the preliminary estimates which ignore the latent structure. We further modify the methodology and theory to allow for structural breaks in either the group membership, group number or group-specific coefficient functions. Numerical studies including Monte-Carlo simulation and an empirical application are presented to examine the finite-sample performance of the developed model and methodology.

AoS Other · 发表于 2026-04-01

Eigenvector overlaps in large sample covariance matrices and nonlinear shrinkage estimators

Zeqin Lin, Guangming Pan

摘要

Consider a data matrix Y=[y1,…,yN] of size M×N, where the columns are independent observations from a random vector y with zero mean and population covariance Σ. Let ui and vj denote the left and right singular vectors of Y, respectively. This study investigates the eigenvector/singular vector overlaps ⟨ui,D1uj⟩, ⟨vi,D2vj⟩ and ⟨ui,D3vj⟩, where Dk are general deterministic matrices with bounded operator norms. In the high-dimensional regime, where the dimension M scales proportionally with the sample size N, we establish the convergence in probability of these eigenvector overlaps towards their deterministic counterparts with explicit convergence rates. Building upon these findings, we offer a more precise characterization of the loss associated with Ledoit and Wolf’s nonlinear shrinkage estimators of the population covariance Σ.

AoS Machine Learning · 发表于 2026-04-01

Distributionally robust learning for multisource unsupervised domain adaptation

Zhenyu Wang, Peter Bühlmann, Zijian Guo

摘要

Empirical risk minimization often performs poorly when the distribution of the target domain differs from those of the source domains. To address such potential distributional shifts, we develop an unsupervised domain adaptation approach that leverages labeled data from multiple source domains and unlabeled data from the target domain. We introduce a distributionally robust model that optimizes an adversarial reward based on explained variance across a class of target distributions, ensuring generalization to the target domain. We show that the proposed robust model is a weighted average of conditional outcome models from the source domains. This formulation allows us to compute the robust model through the aggregation of source models, which can be estimated using various machine learning algorithms of the user’s choice such as random forests, boosting and neural networks. Additionally, we introduce a bias-correction step to obtain a more accurate aggregation weight, which is effective for various machine learning algorithms. Our framework can be interpreted as a distributionally robust federated learning approach that satisfies privacy constraints while providing insights into the importance of each source for prediction on the target domain. The performance of our method is evaluated on both simulated and real data.

AoS Other · 发表于 2026-04-01

Analysis of singular subspaces under random perturbations

Ke Wang

摘要

We present a comprehensive analysis of singular vector and singular subspace perturbations in the signal-plus-noise matrix model with random Gaussian noise. Assuming a low-rank signal matrix, we extend the Davis–Kahan–Wedin theorem in a fully generalized manner, applicable to any unitarily invariant matrix norm, building on previous results by O’Rourke, Vu, and the author. Our analysis provides fine-grained insights, including ℓ∞ bounds for singular vectors, ℓ2,∞ bounds for singular subspaces, and results for linear and bilinear functions of singular vectors. Additionally, we derive ℓ2,∞ bounds on perturbed singular vectors, taking into account the weighting by their corresponding singular values. Finally, we explore practical implications of these results in the Gaussian mixture model and the submatrix localization problem.

AoS Bayesian Methods Nonparametric & Semiparametric · 发表于 2026-04-01

Adaptive Bayesian regression on data with low intrinsic dimensionality

Tao Tang, Nan Wu, Xiuyuan Cheng, David Dunson

摘要

We study how the posterior contraction rate under a Gaussian process (GP) prior depends on the intrinsic dimension of the predictors and the smoothness of the regression function. An open question is whether a generic GP prior that does not incorporate knowledge of the intrinsic lower-dimensional structure of the predictors can attain an adaptive rate for a broad class of such structures. We show that this is indeed the case, establishing conditions under which the posterior contraction rates become adaptive to the intrinsic dimension in terms of the covering number of the data domain (the Minkowski dimension) and prove the nonparametric posterior contraction rate, up to a logarithmic factor. When the domain is a compact manifold, we prove the RKHS approximation to intrinsically defined Hölder functions on the manifold of any order of smoothness by a novel analysis, leading to the optimal adaptive posterior contraction rate. We propose an empirical Bayes prior on the kernel bandwidth using kernel affinity and k-nearest neighbor statistics, bypassing explicit estimation of the intrinsic dimension. The efficiency of the proposed Bayesian regression approach is demonstrated in various numerical experiments.

SMMR Survival Analysis High-Dimensional Nonparametric & Semiparametric · 发表于 2026-03-31

Penalized estimation of linear transformation models for interval-censored data with time-dependent covariates

Minggen Lu, Yahui Zhang, Chin-Shang Li, Guogen Shan

摘要

We investigate efficient estimation strategies for partially linear transformation models with time-dependent covariates under interval censoring. The unknown monotone function is approximated using a monotone B -spline basis to enable flexible semiparametric modeling, and we develop a computationally efficient nested hybrid EM algorithm that integrates Newton’s method with isotonic regression. To support large-sample inference, we propose a straightforward variance–covariance estimation procedure for the regression parameters and introduce a score test to assess the adequacy of the proportional hazards (PH) specification within the broader class of transformation models. The numerical performance of the penalized estimators is examined extensively and compared with both the time-invariant covariate model by Lu et al. and the semiparametric transformation model by Zeng et al. Finally, the proposed methodology is applied to data from the National Alzheimer’s Coordinating Center (NACC) to demonstrate its practical utility in a real-world clinical setting.

Biometrika Other · 发表于 2026-03-31

Parallel computations for Metropolis Markov chains with Picard maps

S Grazzi, G Zanella

摘要

Summary: We develop parallel algorithms for simulating zeroth-order, also known as gradient-free, Metropolis Markov chains based on the Picard map. For random-walk Metropolis Markov chains targeting log-concave distributions $ \pi $ on $ \mathbb{R}^{d} $, our algorithm generates samples close to $ \pi $ in $ \mathcal{O}(\surd{d}) $ parallel iterations using $ \mathcal{O}(\surd{d}) $ processors, thereby speeding up the convergence of the corresponding sequential implementation by a factor $ \surd{d} $. Furthermore, a modification of our algorithm generates samples from an approximate measure $ \pi_{r} $ in $ \mathcal{O}(1) $ parallel iterations and $ \mathcal{O}(d) $ processors. We empirically assess the performance of the proposed algorithms in high-dimensional regression problems, an epidemic model where the gradient is unavailable and a real-word application in precision medicine. Our algorithms are straightforward to implement and may constitute a useful tool for practitioners seeking to sample from a prescribed distribution $ \pi $ using only pointwise evaluations of $ \log\pi $ and parallel computing.

SMMR Causal Inference Networks & Graphs · 发表于 2026-03-31

Improving finite sample performance of causal discovery by exploiting temporal structure

Christine W. Bang, Janine Witte, Ronja Foraita, Vanessa Didelez

摘要

Methods of causal discovery aim to identify causal structures in a data-driven way. Existing algorithms are known to be unstable and sensitive to statistical errors, and are therefore rarely used with biomedical or epidemiological data. We investigate an algorithm that efficiently exploits temporal structure, so-called tiered background knowledge , for estimating causal structures. Tiered background knowledge is readily available from, for example, cohort or registry data. When used efficiently it renders the algorithm more robust to statistical errors and ultimately increases accuracy in finite samples. We describe the algorithm and illustrate how it proceeds. Moreover, we offer formal proofs as well as examples of desirable properties of the algorithm, which we demonstrate empirically in an extensive simulation study. To illustrate its usefulness in practice, we apply the algorithm to data from a children’s cohort study investigating the interplay of diet, physical activity and other lifestyle factors for health outcomes.

SMMR Causal Inference Nonparametric & Semiparametric · 发表于 2026-03-31

Eliminating residual confounding in the stratified estimator via smoothing along with the propensity score

Naoto Tsujimoto, Satoshi Hattori

摘要

The stratified estimator by the propensity score is one of the most popular estimator for the average causal effect in the presence of confounding. Despite of its advantages of robustness and simplicity, it has a serious shortcoming of residual confounding even with the correctly specified propensity score. On the other hand, the inverse probability weighting estimator by the propensity score is free from residual confounding and many corresponding variants have been proposed. Wang et al. (2021, Robust estimation of propensity score weights via subclassification ) pointed it out that the stratified estimator can be regarded as a special case of the inverse probability weighting estimator with piecewise constant propensity score and proposed a method eliminating residual confounding in the stratified estimator. In this paper, we provide an alternative interpretation of the stratified estimator as the outcome regression with a piecewise constant function over the propensity score. By considering kernel smoothing, an estimator free from residual confounding is proposed based on the stratified estimator, which preserves the robustness of the stratified estimator. We also propose a doubly robust estimator, which does not rely on the inverse probability weighting.

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 页