2026 年的期刊文章 · 第 28 页

共 1756 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2026-03-30 到 2026-03-23。

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

JASA Other · 发表于 2026-03-30

Systemic and Systematic Risks-Driven Marginal Expected Side-effect

Liujun Chen, Deyuan Li, Zhengjun Zhang

摘要

System-wide stress events (SWSE) can occur in social, economic, financial, environmental, medical, and many other systems. An SWSE can cause severe damage to some or all members in the system. Understanding and measuring the marginal expected side-effect (MES) for each member caused by an SWSE can lead to better risk management and decision-making. This paper defines an SWSE as a scenario in which either a representative systematic index or the index of the worst-performing member (systemic index) in a system exceeds its respective prescribed threshold. This broadened SWSE definition introduces an innovative systemic and systematic risks-driven marginal expected side-effect (SYS2MES) index. We construct estimators for SYS2MES and establish their asymptotic theories within the multivariate extreme value theory framework, including tail dependent and tail independent scenarios, thereby enhancing its applicability to a wide range of models. The finite sample performance of the estimators is investigated in a simulation study. The advantages of SYS2MES are demonstrated by studying its application to Dow Jones’ 30 stocks and COVID-19 data in the United States, yielding more meaningful results than existing approaches. *(abstract via OpenAlex)*

Biometrics Survival Analysis Latent Variables & Mixtures · 发表于 2026-03-30

Simultaneous clustering and estimation of additive shape invariant models for recurrent event data

Zitong Zhang, Shizhe Chen

摘要

Technological advancements have enabled the recording of spiking activities from large neuron ensembles, presenting an exciting yet challenging opportunity for statistical analysis. This project considers the challenges from a common type of neuroscience experiments, where randomized interventions are applied over the course of each trial. The objective is to identify groups of neurons with unique stimulation responses and estimate these responses. The observed data, however, comprise superpositions of neural responses to all stimuli, which is further complicated by varying response latencies across neurons. We introduce a novel additive shape invariant model that is capable of simultaneously accommodating multiple clusters, additive components, and unknown time-shifts. We establish conditions for the identifiability of model parameters, offering guidance for the design of future experiments. We examine the properties of the proposed algorithm through simulation studies, and apply the proposed method on neural data collected in mice. Specifically, we analyze neural spike trains recorded using Neuropixels probes during a visual discrimination task with two independent randomized stimuli. The analysis identifies three distinct functional groups of neurons that exhibit heterogeneous response patterns to the stimuli, including both stimulus-specific activity and variability in response timing.

SMMR Networks & Graphs Bayesian Methods · 发表于 2026-03-30

Joint estimation of multiple graphical models for an fMRI study of brain connectivity networks

Lizhe Sun, Xiaojuan Han, Aiying Zhang

摘要

Investigating changes and similarities in brain connectivity networks across task conditions is a central topic in neuroscience. We propose a novel framework for jointly estimating multiple graphical models using a hybrid Bayesian integration technique that can handle high-dimensional or large-scale data sets. This framework accommodates multiple graphical models and performs a series of conditional independence tests to infer the underlying network structures. Theoretical justification for consistency is established, and synthetic experiments demonstrate that our approach outperforms existing methods in both accuracy and robustness. We further apply this framework to an functional magnetic resonance imaging study of dynamic functional connectivity among regions of interest during an emotion-processing task. Results reveal that inter- and intra-module interactions involving the subcortical–cerebellum module are reduced during emotion processing compared to shape processing, highlighting this module’s key role in emotional processing.

SiM Causal Inference Survival Analysis · 发表于 2026-03-30,收录于 2026-05-01

Illustrating Implications of Misaligned Causal Questions and Statistics in Settings With Competing Events and Interest in Treatment Mechanisms

Takuya Kawahara, Sean McGrath, Jessica G. Young

摘要

In the presence of competing events, many investigators are interested in a direct treatment effect on the event of interest that does not capture treatment effects on competing events. Classical survival analysis methods that treat competing events like censoring events, at best, target a controlled direct effect: the effect of the treatment under a difficult to imagine and typically clinically irrelevant scenario where competing events are somehow eliminated. A separable direct effect, quantifying the effect of a future modified version of the treatment, is an alternative direct effect notion that may better align with an investigator's underlying causal question. In this paper, we provide insights into the implications of naively applying an estimator constructed for a controlled direct effect (i.e., “censoring by competing events”) when the actual causal effect of interest is a separable direct effect. We illustrate the degree to which controlled and separable direct effects may take different values, possibly even different signs, and the degree to which these two different effects may be differentially impacted by violation and/or near violation of their respective identifying conditions under a range of data generating scenarios. Finally, we provide an empirical comparison of inverse probability of censoring weighting to an alternative weighted estimator specifically structured for a separable effect using data from a randomized trial of estrogen therapy and prostate cancer mortality.

JASA Optimization & Learning Theory · 发表于 2026-03-30,收录于 2026-09-10

Contextual Dynamic Pricing: Algorithms, Optimality, and Local Differential Privacy Constraints

Zifeng Zhao, Feiyu Jiang, Yi Yu

摘要

(sourced from arXiv preprint 2406.02424v3 — same paper; tandfonline was 403-blocked): "We study contextual dynamic pricing problems where a firm sells products to T sequentially-arriving consumers, behaving according to an unknown demand model. The firm aims to minimize its regret over a clairvoyant that knows the model in advance. The demand follows a generalized linear model (GLM), allowing for stochastic feature vectors in ℝ^d encoding product and consumer information. We first show the optimal regret is of order √(dT), up to logarithmic factors, improving existing upper bounds by a √d factor. This optimal rate is materialized by two algorithms: a confidence bound-type algorithm and an explore-then-commit (ETC) algorithm. A key insight is an intrinsic connection between dynamic pricing and contextual multi-armed bandit problems with many arms with a careful discretization. We further study contextual dynamic pricing under local differential privacy (LDP) constraints. We propose a stochastic gradient descent-based ETC algorithm achieving regret upper bounds of order d√(T)/ε, up to logarithmic factors, where ε > 0 is the privacy parameter. The upper bounds with and without LDP constraints are matched by newly constructed minimax lower bounds, characterizing costs of privacy. Moreover, we extend our study to dynamic pricing under mixed privacy constraints, improving the privacy-utility tradeoff by leveraging public data. This is the first time such setting is studied in the dynamic pricing literature and our theoretical results seamlessly bridge dynamic pricing with and without LDP.

JRSSB Spatial & Time Series · 发表于 2026-03-30

Beyond the mean: limit theory and tests for infinite-mean autoregressive conditional durations

Giuseppe Cavaliere, Thomas Mikosch, Anders Rahbek, Frederik Vilandt

摘要

Integrated autoregressive conditional duration (ACD) models serve as counterparts to integrated generalized autoregressive conditional heteroskedastic models used for financial returns. However, despite their resemblance, asymptotic theory for ACD is still incomplete. Central challenges arise from the facts that (i) integrated ACD processes imply durations with infinite expectation and (ii) conventional asymptotic approaches break down due to the randomness in the number of durations within a fixed observation period. We fill this gap in the literature and provide a unified asymptotic theory for the (quasi)maximum likelihood estimator for integrated ACD models. Based on the new theoretical results, we also provide a novel framework for hypothesis testing in duration models, enabling inference on a key empirical question: whether durations possess a finite or infinite expectation. We apply our results to high-frequency cryptocurrency exchange traded fund (ETF) trading data. Motivated by parameter estimates near the integrated ACD region, we assess whether durations between trades in these markets have finite expectation, an assumption often made implicitly in the literature on point process models. Our empirical findings indicate infinite-mean durations for all five cryptocurrency ETFs examined, and we reject the integrated ACD hypothesis in favour of heavier-tailed alternatives for four out of the five ETFs.

Biometrics Genetics & Omics · 发表于 2026-03-28

Rank-adaptive covariance testing with applications to genomics and neuroimaging

David Veitch, Yinqiu He, Jun Young Park

摘要

In biomedical studies, testing for differences in covariance may offer scientific insights, especially when differences are driven by complex joint behavior between features. However, when differences in joint behavior are weakly dispersed across many dimensions and arise from differences in low-rank structures within the data, as is often the case in genomics and neuroimaging, existing two-sample covariance testing methods may suffer from power loss. The Ky-Fan(k) norm, defined by the sum of the top k singular values, is a simple and intuitive matrix norm able to capture signals caused by differences in low-rank structures between matrices, but its statistical properties in hypothesis testing have not been studied well. In this paper, we investigate the behavior of the Ky-Fan(k) norm in two-sample covariance testing. Ultimately, we propose a novel methodology, rank-adaptive covariance testing (RACT), which is able to leverage differences in low-rank structures found in the covariance matrices of two groups in order to maximize power. RACT uses permutation for statistical inference, ensuring an exact Type I error control. We validate RACT in simulation studies and evaluate its performance when testing for differences in gene expression networks between two types of lung cancer, as well as testing for covariance heterogeneity in diffusion tensor imaging data taken on two different scanner types.

Biometrika Spatial & Time Series Survival Analysis · 发表于 2026-03-28

Palm distributions of superposed point processes for statistical inference

M Beraha, F Camerlenghi, L Ghilotti

摘要

Palm distributions play a central role in the study of point processes and their associated summary statistics. In this work, we characterize the Palm distributions of the superposition of independent point processes, establishing a simple mixture representation depending on the point processes’ Palm distributions and moment measures. We explore two statistical applications enabled by our main result. First, we consider minimum contrast estimation for corrupted point processes. Second, we investigate the class of shot noise Cox processes and derive explicit expressions for their higher-order Palm distributions. In the finite case, we further obtain a tractable expression for the Janossy density, which plays the role of a likelihood function and thus can be used for new likelihood-based inference strategies. Extensions to the superposition of multiple point processes and to higher-order Palm distributions are also presented.

JRSSB Causal Inference · 发表于 2026-03-28

Double cross-fit doubly robust estimators: Beyond series regression

Alec McClean, Sivaraman Balakrishnan, Edward H Kennedy, Larry Wasserman

摘要

Double cross-fit doubly robust (DCDR) estimators, which train nuisance function estimators on separate samples, are effective new estimators for causal functionals. We establish several novel theoretical results for them, building on recent work. We provide a structure-agnostic error analysis, which holds with generic nuisance functions and estimators. Then, we propose n-consistent DCDR estimators with undersmoothed local polynomial regression and k-Nearest Neighbours and a minimax rate-optimal DCDR estimator with undersmoothed kernel regression. Finally, we demonstrate inference is possible even in the non-root-n regime with a central limit theorem for an undersmoothed DCDR estimator. We reinforce our theoretical results with simulation experiments.

Biometrika Bayesian Methods · 发表于 2026-03-28

Conditioning on posterior samples for flexible frequentist goodness-of-fit testing

Ritwik Bhaduri, Aabesh Bhattacharyya, Rina Foygel Barber, Lucas Janson

摘要

Summary: Tests of goodness of fit are used in nearly every domain where statistics is applied. One powerful and flexible approach is to sample artificial datasets that are exchangeable with the real data under the null hypothesis (but not under the alternative), as this allows the analyst to conduct a valid test using any test statistic they desire. Such sampling is typically done by conditioning on either an exact or an approximate sufficient statistic, but existing methods for doing so have significant limitations, which either preclude their use or substantially reduce their power or computational tractability for many important models. In this article, we propose to condition on samples from a Bayesian posterior distribution, which constitute a very different type of approximate sufficient statistic than those considered in previous work. Our approach, termed approximately co-sufficient sampling via Bayes, considerably expands the scope of thisflexible type of goodness-of-fit testing. We prove the approximate validity of the resulting test and demonstrate its utility on three common null models to which no existing methods apply, as well as its superior performance on models to which existing methods do apply.

Biometrics Clinical Trials · 发表于 2026-03-27

Making all pairwise comparisons in multi-arm clinical trials without control treatment

T Burnett, T Jaki

摘要

The standard paradigm for confirmatory clinical trials is to compare experimental treatments with a control, for example, the standard of care or a placebo. However, it is not always the case that a suitable control exists. Efficient statistical methodology is well studied in the setting of randomized controlled trials. This is not the case if one wishes to compare several experimental treatments with no control arm. We propose hypothesis testing methods suitable for use in such a setting. These methods are efficient, ensuring the error rate is controlled at exactly the desired rate with no conservatism. This in turn yields an improvement in power when compared with standard methods one might otherwise consider using, such as a Bonferroni adjustment. The proposed testing procedure is also highly flexible. We show how it may be extended for use in multistage adaptive trials, covering the majority of scenarios in which one might consider the use of such procedures in the clinical trials setting. With such a highly flexible nature, these methods may also be applied more broadly outside of a clinical trials setting.

JASA Networks & Graphs Machine Learning Genetics & Omics · 发表于 2026-03-27

Heterogeneous Gene Network Estimation for Single-Cell Transcriptomic Data via a Joint Regularized Deep Neural Network

Jingyuan Yang, Tao Li, Tianyi Wang, Shuangge Ma, Mengyun Wu

摘要

Estimation of intracellular gene networks has been a critical component of single-cell transcriptomic data analysis, which can provide crucial insights into the complex interplay between genes, facilitating the discovery of the biological basis of human life at single-cell resolution. Despite notable achievements, existing methodologies often falter in their practicality, primarily due to their narrow focus on simplistic linear relationships and inadequate handling of cellular heterogeneity. To bridge these gaps, we propose a joint regularized deep neural network method incorporating Mahalanobis distance-based K-means clustering (JRDNN-KM) to estimate multiple networks for various cell subgroups simultaneously, accounting for both unknown cellular heterogeneity and zero inflation, and, more importantly, complex nonlinear relationships among genes. We introduce an innovative selection layer for network construction, along with hidden layers that include both shared and subgroup-specific neurons, to capture common patterns and subgroup-specific variations across networks. Applied to real single-cell transcriptomic data from multiple tissues and species, JRDNN-KM demonstrates higher accuracy and biological interpretability in network estimation, and more accurately identifies cell subgroups compared to current state-of-the-art methods. Building on network construction, we further find hub genes with important biological implications and modules with statistical enrichment of biological processes. *(abstract via OpenAlex)*

JASA Networks & Graphs Latent Variables & Mixtures · 发表于 2026-03-27

Efficient Analysis of Latent Spaces in Heterogeneous Networks

Yuang Tian, Jiajin Sun, Yinqiu He

摘要

This work proposes a unified framework for efficient estimation under latent space modeling of heterogeneous networks. We consider a class of latent space models that decompose latent vectors into shared and network-specific components across networks. We develop a novel procedure that first identifies the shared latent vectors and further refines estimates through efficient score equations to achieve statistical efficiency. Oracle error rates for estimating the shared and heterogeneous latent vectors are established simultaneously. The analysis framework offers remarkable flexibility, accommodating various types of edge weights under general distributions. *(abstract via OpenAlex)*

SMMR Survival Analysis Testing & Uncertainty Nonparametric & Semiparametric · 发表于 2026-03-27

Covariate hypothesis tests for the cure rate in mixture cure models based on martingale difference correlation

Blanca E Monroy-Castillo, María Amalia Jácome, Ricardo Cao, Ingrid Van Keilegom

摘要

Cure models are a class of survival models used to analyze time-to-event data that allow the possibility that the event never occurs for a certain percentage 1 − p , of the population. These methods allow for direct modelling of the cure rate and the influence of covariates on this rate. A common goal is to test whether the cure rate depends on a specific covariate or a set of covariates. The availability of methods to test the effect of covariates on the cure rate is limited in the literature. This paper proposes nonparametric hypothesis tests for the effect of covariates on the cure probability based on the martingale difference correlation. Two methods are used to approximate the null distribution of the test statistic: A permutation test and a chi-square test. The methodology is further extended to the case of two covariates using the partial martingale difference correlation. The performance of the proposed tests is evaluated through a simulation study under various scenarios, and the method is applied to a dataset on rheumatoid arthritis patients.

JASA Causal Inference Bayesian Methods Spatial & Time Series · 发表于 2026-03-27

Bayesian Image Mediation Analysis

Yuliang Xu, Timothy D. Johnson, Mary Heitzeg, Jian Kang

摘要

Mediation analysis aims to separate the indirect effect through mediators from the direct effect of the exposure on the outcome. It is challenging to perform mediation analysis with neuroimaging data which involves high dimensionality, complex spatial correlations, sparse activation patterns and relatively low signal-to-noise ratio. To address these issues, we develop a new spatially varying coefficient structural equation model for Bayesian Image Mediation Analysis (BIMA). We define spatially varying mediation effects within the potential outcomes framework, employing a soft-thresholded Gaussian process prior for functional parameters. We establish posterior consistency for spatially varying mediation effects along with selection consistency on important regions that contribute to the mediation effects. We develop an efficient posterior computation algorithm scalable to analysis of large-scale imaging data. Through extensive simulations, we show that BIMA can improve the estimation accuracy and computational efficiency for high-dimensional mediation analysis over existing methods. We apply BIMA to analyze behavioral and fMRI data in the Adolescent Brain Cognitive Development (ABCD) study with a focus on inferring the mediation effects of the parental education level on the children's general cognitive ability that are mediated through the working memory brain activity. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via PubMed)*

JASA Genetics & Omics · 发表于 2026-03-27

Analyzing Cross-Trait Genetic Architecture with the BIGA Cloud Computing Platform

Yujue Li, Fei Xue, Bingxuan Li, Yilin Yang, Zirui Fan, Juan Shu, Xiaochen Yang, Xiyao Wang, Jinjie Lin, Carlos Copana, Bingxin Zhao

摘要

As large-scale biobanks provide increasing access to deep phenotyping and genomic data, genome-wide association studies (GWAS) are rapidly uncovering the genetic architecture behind various complex traits and diseases. GWAS publications typically make their summary-level data (GWAS summary statistics) publicly available, enabling further exploration of genetic overlaps between phenotypes gathered from different studies and cohorts. However, systematically analyzing high-dimensional GWAS summary statistics for thousands of phenotypes can be both logistically challenging and computationally demanding. In this paper, we introduce BIGA (https://bigagwas.org/), a website that aims to offer unified data analysis pipelines and processed data resources for cross-trait genetic architecture analyses using GWAS summary statistics. We have developed a framework to implement statistical genetics tools on a cloud computing platform, combined with extensive curated GWAS data resources. Through BIGA, users can upload data, submit jobs, and share results, providing the research community with a convenient tool for consolidating GWAS data and generating new insights. *(abstract via PubMed)*

SMMR Diagnostics & Prediction Missing Data & Measurement Error · 发表于 2026-03-27

Addressing nonignorable missing data and heterogeneity in prognostic biomarker assessment

Xinran Huang, Ruosha Li, Jing Ning, for the Alzheimer’s Disease Neuroimaging Initiative

摘要

Covariate-specific and time-dependent area-under-curve (AUC) is often used to evaluate the discriminative performance of biomarkers with time-to-event outcomes, particularly when certain covariates influence biomarkers’ accuracy. In biomarker research, despite extensive efforts, missing data remain unavoidable, with nonignorable missingness posing significant challenges. This article focuses on estimating the impact of covariates on time-dependent AUC in the presence of nonignorable missing biomarkers. Assuming a parametric model on the missing probability, we leverage instrumental variables to address the identifiable issue and estimate the unknown parameters. We integrate the inverse probability weighting approach into the score equation of a pseudo partial likelihood for estimation and inference. Additionally, we establish the asymptotic properties of the proposed estimators. Through simulation studies, we evaluate finite sample performance of the proposed estimators, and apply the method to data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database.

Biometrics Spatial & Time Series Optimization & Learning Theory Genetics & Omics · 发表于 2026-03-26

LLOT: application of Laplacian Linear Optimal Transport in spatial transcriptome reconstruction

Junhao Zhu, Kevin Zhang, Dehan Kong, Zhaolei Zhang

摘要

Single-cell RNA sequencing (scRNA-seq) allows transcriptional profiling, and cell-type annotation of individual cells. However, sample preparation in typical scRNA-seq experiments often homogenizes the samples, thus spatial locations of individual cells are often lost. Although spatial transcriptomic techniques, such as in situ hybridization (ISH) or Slide-seq, can be used to measure gene expression in specific locations in samples, it remains a challenge to measure or infer expression level for every gene at a single-cell resolution in every location in tissues. Existing computational methods show promise in reconstructing these missing data by integrating scRNA-seq data with spatial expression data such as those obtained from spatial transcriptomics. Here we describe Laplacian Linear Optimal Transport (LLOT), an interpretable method to integrate single-cell and spatial transcriptomics data to reconstruct missing information at a whole-genome and single-cell resolution. LLOT iteratively corrects platform effects and employs Laplacian Optimal Transport to decompose each spot in spatial transcriptomics data into a spatially-smooth probabilistic mixture of single cells. We benchmark LLOT against several existing methods on multiple datasets from different measurement technologies, including in situ hybridization, Slide-seq, 10x Visium, and Visium HD. The results demonstrate that LLOT provides an interpretable and versatile framework for reconstructing spatial gene expression and inferring cell locations.

JASA Latent Variables & Mixtures · 发表于 2026-03-26

Conditional Partial Exchangeability: A Probabilistic Framework for Multi-View Clustering

Beatrice Franzolini, Maria De Iorio, Johan Eriksson

摘要

Standard clustering techniques assume a common clustering configuration for all features in a dataset. However, when dealing with multi-view or longitudinal data, the clusters’ number, frequencies, and shapes may need to vary across features to accurately capture dependence structures and heterogeneity. In this setting, classical model-based clustering fails to account for within-subject dependence across domains. We introduce conditional partial exchangeability, a novel probabilistic paradigm for dependent random partitions of the same objects across distinct domains. Additionally, we study a wide class of Bayesian clustering models based on conditional partial exchangeability, which allows for flexible dependent clustering of individuals across features, capturing the specific contribution of each feature and the within-subject dependence while ensuring computational feasibility. *(abstract via OpenAlex)*

JASA Longitudinal Data · 发表于 2026-03-25

Subtype-Aware Registration of Longitudinal Electronic Health Records

Xin Gai, Shiyi Jiang, Anru R. Zhang

摘要

Electronic Health Records (EHRs) contain extensive patient information that can inform downstream clinical decisions, such as mortality prediction, disease phenotyping, and disease onset prediction. A key challenge in EHR data analysis is the temporal gap between when a condition is first recorded and its actual onset time. Such timeline misalignment can lead to artificially distinct biomarker trends among patients with similar disease progression, undermining the reliability of downstream analyses and complicating tasks such as disease subtyping and outcome prediction. To address this challenge, we provide a subtype-aware timeline registration method that leverages data projection and discrete optimization to correct timeline misalignment. Through simulation and real-world data analyses, we demonstrate that the proposed method effectively aligns distorted observed records with the true disease progression patterns, enhancing subtyping clarity and improving performance in downstream clinical analyses. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*

JASA Spatial & Time Series · 发表于 2026-03-25,收录于 2026-09-24

Structural Identification for Spatio-Temporal Dynamic Models

Not retrieved — tandfonline page blocked; no matching preprint located

摘要

Identifying latent cluster structures in spatial trends constitutes an important yet challenging task in diverse applications. In this paper, we propose a novel method based on a discrepancy measure over small spatial blocks that effectively uncovers heterogeneity within dynamic spatial data. Our approach effectively detects boundaries where structural changes occur, thus allowing for more nuanced insights into underlying spatial patterns. Unlike methods predicated on strong stationarity assumptions, our framework accommodates piecewise-defined parameters and irregular sampling locations, enabling its broad applicability to real-world datasets. We further establish asymptotic properties and limit distributions of the proposed methods by leveraging the notion of spatial physical dependence, accounting for correlations across spatial domains. Simulations and real data analyses confirm the effectiveness of the method, highlighting its robustness and accuracy in identifying complex spatio-temporal structures. *(abstract via OpenAlex)*

SMMR Other · 发表于 2026-03-25

Randomization and allocation procedures for master protocol trials of single-arm studies

Peter Jacko, Günter Heimann, Tom Parke

摘要

In this paper we propose and examine randomization and allocation procedures in master protocol trials where each subtrial is performed as a single-arm study, which is motivated mainly by rare diseases. The subtrial analysis is done by comparing the single-arm observations data to historical data or expert judgment and following prespecified decision rules, without using data from any other study in the master protocol trial. The objective of this paper is to examine relevant options for participant allocation procedures in a variety of patterns in which the subtrials enter the master protocol trial, such as umbrella, platform and perpetual patterns. Our simulation study shows that our proposed allocation procedures using the predictive probability of current study success may bring substantial efficiency gains in the master protocol trial, making a clever allocation of the in-trial participants to achieve study success declarations sooner, benefiting the out-of-trial population.

JASA Other · 发表于 2026-03-25

Possibilistic Inferential Models: A Review

Ryan Martin

摘要

An inferential model (IM) is a model describing the construction of provably reliable, data-driven uncertainty quantification and inference about relevant unknowns. IMs and Fisher’s fiducial argument have similar objectives, but a fundamental distinction between the two is that the former doesn’t require that uncertainty quantification be probabilistic, offering greater flexibility and allowing for a proof of its reliability. Important recent developments have been made thanks in part to newfound connections with the imprecise probability literature, in particular, possibility theory. The brand of possibilistic IMs studied here are straightforward to construct, have very strong frequentist-like reliability properties, and offer fully conditional, Bayesian-like (imprecise) probabilistic reasoning. This paper reviews these key recent developments, describing the new theory, methods, and computational tools. A generalization of the basic possibilistic IM is also presented, making new and unexpected connections with ideas in modern statistics and machine learning, e.g., bootstrap and conformal prediction. *(abstract via OpenAlex)*

JASA Bayesian Methods · 发表于 2026-03-25

Optimal Plug-in Gaussian Processes for Modeling Derivatives

Zejian Liu, Meng Li

摘要

Derivatives are a key nonparametric functional in wide-ranging applications where the rate of change of an unknown function is of interest. In the Bayesian paradigm, Gaussian processes (GPs) are routinely used as a flexible prior for unknown functions, and are arguably one of the most popular tools in many areas. However, little is known about the optimal modeling strategy and theoretical properties when using GPs for derivatives. In this article, we study a plug-in strategy by differentiating the posterior distribution with GP priors for derivatives of any order. This practically appealing plug-in GP method has been previously perceived as suboptimal and degraded, but this is not necessarily the case. We provide posterior contraction rates for plug-in GPs and establish that they achieve optimal rates simultaneously for all derivative orders. We show that the posterior measure of the regression function and its derivatives, with the same choice of hyperparameter that does not depend on the order of derivatives, converges at the minimax optimal rate up to a logarithmic factor for functions in certain classes. We analyze a data-driven hyperparameter tuning method based on empirical Bayes, and show that it satisfies the optimal rate condition while maintaining computational efficiency. This article to the best of our knowledge provides the first positive result for plug-in GPs in the context of inferring derivative functionals, and leads to a practically simple nonparametric Bayesian method with optimal and adaptive hyperparameter tuning for simultaneously estimating the regression function and its derivatives. Simulations show competitive finite sample performance of the plug-in GP method. A climate change application for analyzing the global sea-level rise is discussed. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*

JASA Testing & Uncertainty Nonparametric & Semiparametric · 发表于 2026-03-25

Nonparametric Bootstrap Inference for the Eigenvalues of Geophysical Tensors

Kassel L. Hingee, Janice L. Scealy, Andrew T. A. Wood

摘要

Symmetric matrices (tensors) are measured in geophysics and other disciplines, including in medical imaging, and typically their eigenvalues have valuable scientific interpretations. We design pivotal bootstrap hypothesis tests of specified eigenvalues or eigenvalue multiplicities in one-sample situations and for equal eigenvalues in k-sample situations. Our tests are more broadly applicable than existing tests by allowing very general distributions, allowing three or more samples, and accounting for common constraints in geophysical measurements. Simulations indicate that our tests generally perform well and have improved power in situations where there are existing formal hypothesis tests (eigenvalue multiplicity tests and 2-sample tests of unconstrained eigenvalues). We show fast O(n−2) convergence of test size for our pivotal k-sample tests. We also propose confidence regions for eigenvalues and apply our tests to four geophysical data sets. An R package accompanies this article. *(abstract via OpenAlex)*

JASA Networks & Graphs · 发表于 2026-03-25

Multivariate Analysis for Multiple Network Data via Semi-Symmetric Tensor PCA

Michael Weylandt, George Michailidis

摘要

Network data are commonly collected in a variety of applications, representing either directly measured or statistically inferred connections between subjects or features of interest. In an increasing number of domains, these networks are collected over time, such as repeated interactions between users of a social media platform, or across multiple subjects, such as in multi-subject neuroimaging studies. When analyzing multiple large networks, dimensionality reduction techniques are often used to embed networks in a more tractable low-dimensional space. To this end, we develop a framework for principal components analysis (PCA) on collections of networks via a specialized tensor decomposition, termed Semi-Symmetric Tensor PCA or SST-PCA, and analyze it theoretically. Notably, we show that SST-PCA achieves the same accuracy as classical matrix PCA, with error proportional to the square root of the number of vertices and not the number of edges as might be expected. Our framework inherits many of the strengths of classical PCA and is suitable for a wide range of unsupervised learning tasks, including identifying principal networks, isolating changepoints and outliers, and for characterizing the “variability network” of the most varying edges. Finally, we demonstrate the effectiveness of SST-PCA in simulation and on an example from empirical social studies. *(abstract via OpenAlex)*

JASA Other · 发表于 2026-03-25

Factor Augmented Matrix Regression

Elynn Chen, Jianqing Fan, Xiaonan Zhu

摘要

We introduce Factor-Augmented Matrix Regression (FAMAR) to address the growing applications of matrix-variate data and their associated challenges, particularly with high-dimensionality and covariate correlations. FAMAR encompasses two key algorithms. The first is a novel non-iterative approach that efficiently estimates the factors and loadings of the matrix factor model, using techniques of pre-training, diverse projection, and block-wise averaging. The second algorithm offers an accelerated solution for penalized matrix factor regression. Both algorithms are supported by established statistical and numerical convergence properties. Empirical evaluations conducted on synthetic and real economics datasets demonstrate FAMAR’s superiority in terms of accuracy, interpretability, and computational speed. An application to economic data showcases how matrix factors can be incorporated to predict the GDPs of the countries of interest and the influence of these factors on the GDPs. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*

JASA Other · 发表于 2026-03-25

Estimating Multiple Structural Breaks in Large Panels With Unobserved Heterogeneity

Wenxin Huang, Yucheng Sun, Yiru Wang

摘要

This article proposes a novel algorithm to identify potentially multiple breaks in linear panel data models, while the slope coefficients can have individual heterogeneity and the cross-sectional dimension is allowed to diverge jointly with the time span. The algorithm adopts a shrinkage penalty to detect breaks, and delivers a consistent estimator for the number of breaks and fraction-consistent estimators for break dates. In the presence of latent grouped heterogeneity, we employ the classifier-Lasso method to further derive super-consistent break-date estimators based on the break identification result given by the shrinkage algorithm. We demonstrate the satisfactory performance of the proposed methodology in finite samples with Monte Carlo simulations. Empirically, we illustrate the use of our methods to study structural breaks and unobserved heterogeneity in cross-sectional risk premiums. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*

JASA Other · 发表于 2026-03-25

Emerging Knowledge Trend in Statistical Research: A Content-Based Analysis Using Covariate-Assisted Dynamic Topic Model

Chenxuan He, Feifei Wang, Liping Zhu

摘要

Exploring the emerging knowledge trends within a particular discipline is of great interest to scientific researchers. It helps researchers to understand the historical development of their disciplines and to guide their future research directions. In this work, we focus on the fast-developing discipline, statistics, to investigate its knowledge trend and emerging topics in statistical research. To this end, we collect publications in top-tier statistical journals and statistical-related conferences, and develop a covariate-assisted dynamic topic model (CDTM). It captures the dynamic evolution of topics in statistical publications and also finds the time-varying effects of covariates on topic discussions. To estimate CDTM, a variational inference procedure is applied. The theoretical properties are studied and finite-sample performance is evaluated through simulation experiments. Last, we apply CDTM to the collected academic data. We highlight the advantages of CDTM over other alternative methods and uncover topics that characterize the evolution of statistical knowledge over the past four decades. We also observe that certain research topics in statistics are increasingly aligning with advancements in artificial intelligence. Based on these findings, we gain valuable insights into the historical progression of statistical research, which enables us to better anticipate future trends and guide innovation in the field. *(abstract via OpenAlex)*

JASA Meta-Analysis & Data Integration · 发表于 2026-03-25

Differentially Private Sliced Inverse Regression in the Federated Paradigm

Shuaida He, Jiarui Zhang, Xin Chen

摘要

Sliced inverse regression (SIR), which includes linear discriminant analysis (LDA) as a special case, is a popular and powerful dimension reduction tool. In this article, we extend SIR to address the challenges of decentralized data, prioritizing privacy and communication efficiency. Our approach, termed as federated sliced inverse regression (FSIR), facilitates distributed computing of the sufficient dimension reduction subspace among multiple clients, solely sharing local estimates to protect sensitive datasets from exposure. To guard against potential adversary attacks, FSIR employs diverse perturbation strategies, including a novel vectorized Gaussian mechanism that guarantees (ε,δ)-differential privacy at a low cost of statistical accuracy. Additionally, FSIR achieves a tight composition of various privacy mechanisms by adopting a hypothesis testing perspective on differential privacy. It also incorporates a collaborative feature screening procedure, enabling effective handling of high-dimensional client data with varying feature sets. Theoretical properties of FSIR are established for both low-dimensional and high-dimensional settings, supported by extensive numerical experiments and real data analysis. *(abstract via OpenAlex)*

JASA Testing & Uncertainty Nonparametric & Semiparametric · 发表于 2026-03-25

Differentially Private Permutation Tests

Ilmun Kim, Antonin Schrab

摘要

Recent years have witnessed growing concerns about the privacy of sensitive data. In response to these concerns, differential privacy has emerged as a rigorous framework for privacy protection, gaining widespread recognition in both academic and industrial domains. While substantial progress has been made in private data analysis, existing methods often suffer from impracticality or a significant loss of statistical efficiency. This article aims to alleviate these concerns in the context of hypothesis testing by introducing differentially private permutation tests. The proposed framework extends classical non-private permutation tests to private settings, maintaining both finite-sample validity and differential privacy in a rigorous manner. The power of the proposed test depends on the choice of a test statistic, and we establish general conditions for consistency and non-asymptotic uniform power. To demonstrate the utility and practicality of our framework, we focus on reproducing kernel-based test statistics and introduce differentially private kernel tests for two-sample and independence testing: dpMMD and dpHSIC. The proposed kernel tests are straightforward to implement, applicable to various types of data, and attain minimax optimal power across different privacy regimes. Our empirical evaluations further highlight their competitive power under various synthetic and real-world scenarios, emphasizing their practical value. The code is publicly available to facilitate the implementation of our framework. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*

JASA Machine Learning Networks & Graphs · 发表于 2026-03-25

Belted and Ensembled Neural Network for Linear and Nonlinear Sufficient Dimension Reduction

Yin Tang, Bing Li

摘要

We introduce a unified, flexible, and easy-to-implement framework of sufficient dimension reduction that can accommodate both linear and nonlinear dimension reduction, and both the conditional distribution and the conditional mean as the targets of estimation. This unified framework is achieved by a specially structured neural network—the Belted and Ensembled Neural Network (BENN)—that consists of a narrow latent layer, which we call the belt, and a family of transformations of the response, which we call the ensemble. By strategically placing the belt at different layers of the neural network, we can achieve linear or nonlinear sufficient dimension reduction, and by choosing the appropriate transformation families, we can achieve dimension reduction for the conditional distribution or the conditional mean. Moreover, thanks to the advantage of the neural network, the method is very fast to compute, overcoming a computation bottleneck of the traditional sufficient dimension reduction estimators, which involves the inversion of a matrix of dimension either p or n. We develop the algorithm and convergence rate of our method, compare it with existing sufficient dimension reduction methods, and apply it to two data examples. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*

SiM Latent Variables & Mixtures Causal Inference Longitudinal Data · 发表于 2026-03-25,收录于 2026-05-01

A Latent‐factor MCACE Model for Multidimensional Outcomes and Treatment Noncompliance with Application to a Longitudinal Trial of Arthritis Health Journal

Lulu Guo, Yi Qian, Diane Lacaille, Hui Xie

摘要

Real‐world randomized controlled trials (RCTs) evaluating multifaceted interventions often employ multiple study outcomes to measure treatment effects on a small set of underlying constructs. Motivated by a longitudinal RCT evaluating a behavioural intervention, the Arthritis Health Journal (AHJ), we propose a latent‐factor multivariate complier average causal effects (MCACE) model for multidimensional longitudinal outcomes with principal strata of compliance types for parsimonious estimation of intervention effects in RCTs with treatment noncompliance. Within each compliance type, a factor regression model relates multiple outcomes to latent constructs, which follow hierarchical regression models. Under the model, high dimensional outcomes are reduced to low dimensional latent factors. This dimension reduction leads to a parsimonious and efficient test of overall CACEs on multiple outcomes, mitigating the multiple testing issues associated with multidimensional outcomes and weak instrumental variable problems associated with low compliance rates. Simulation studies demonstrate that the latent‐factor MCACE model outperforms univariate CACE analysis in terms of both statistical power and Type I error control. The application to the AHJ study selects two underlying factors (self‐efficacy and interaction with health care providers). Significant and beneficial treatment effects are detected on both factors. Overall, our analysis directly answers the main scientific questions posed by the RCT and yields novel findings not discovered previously.

SMMR Survival Analysis Bayesian Methods Missing Data & Measurement Error · 发表于 2026-03-25

A Bayesian transformation model for informative partly interval-censored data with covariates subject to measurement error

Jingjing Jiang, Chunjie Wang

摘要

Linear transformation models are one of the commonly used models for regression analysis of failure time data due to their flexibility. Although the existing literature provides many methods for fitting transformation models with fixed covariates and non-informative censoring, extending these methods to scenarios with covariates subject to measurement error and informative censoring remains challenging. As pointed out in the literature, failure to account for covariate measurement errors or informative censoring may lead to estimation bias or misleading conclusions. Therefore, in this article, we consider a more complicated and general situation where both covariate measurement errors and informative censoring, or more especially informative partly interval censoring, exist. For this problem, we propose a new joint model for regression analysis of such data and present a general Bayesian estimation procedure that can handle both non-informative censoring and informative censoring, using I-splines to approximate unknown functions. To implement this method, we propose a flexible and stable Markov chain Monte Carlo (MCMC) algorithm through a four-stage data augmentation. This method is simple and easy to use. We conduct extensive simulation studies to compare the naive method with the Bayesian method, verifying the effectiveness of the Bayesian method. We also present a practical application to illustrate the proposed method.

Biometrika Bayesian Methods · 发表于 2026-03-24

Sequential Gibbs posteriors with applications to principal component analysis

Steven Winter, Omar Melikechi, David B Dunson

摘要

Summary: Gibbs posteriors are proportional to a prior distribution multiplied by an exponentiated loss function, with a key tuning parameter that weights the information in the loss relative to the prior and provides control of posterior uncertainty. Gibbs posteriors provide a principled framework for likelihood-free Bayesian inference; however, in many situations, the inclusion of a single tuning parameter inevitably leads to poor uncertainty quantification. In particular, regardless of the value of the parameter, credible regions are far from attaining nominal frequentist coverage, even in large samples. We propose a sequential extension to Gibbs posteriors to address this problem. We prove that the proposed sequential posterior exhibits concentration and satisfies a Bernstein–von Mises theorem, which holds under easily verifiable conditions in Euclidean space and on manifolds. As a by-product, we obtain the first Bernstein–von Mises theorem for traditional likelihood-based Bayesian posteriors on manifolds. All methods are illustrated with an application to principal component analysis.

SiM Survival Analysis Nonparametric & Semiparametric Causal Inference · 发表于 2026-03-24,收录于 2026-04-24

Hypothesis Tests of Direct and Indirect Effects Under Various Semicompeting Risks Models

Jih‐Chang Yu, Yen‐Tsung Huang

摘要

The semicompeting risks problem involves two time‐to‐event outcomes where the intermediate outcome may be censored by the primary outcome but not vice versa. It has been shown that semicompeting risks can be formulated as a mediation model, where both the direct effect (DE) and the indirect effect (IE) are studied. This article proposes unified testing procedures based on current causal approaches to evaluate DE and IE under three classic semicompeting risks models: the Clayton copula, gamma frailty, and multistate models. We study the correspondence of the DE and IE with the model parameters and establish testing rules for the two effects under the three models. For statistical inference, we use the U‐statistic approach for the Clayton copula model and nonparametric maximum likelihood estimation for the multistate and gamma frailty models. The simulation study shows that among the three models, the Clayton copula model attains the best statistical power if the model assumption holds but has the potential bias caused by model misspecification; the gamma frailty model is the most robust model by sacrificing the efficiency; the multistate model balances the efficiency and robustness. We apply the proposed method to a hepatitis study, and the aforementioned models unanimously suggest that both hepatitis B and C lead to a higher incidence of liver cancer by increasing liver cirrhosis incidence.

SiM Causal Inference Testing & Uncertainty · 发表于 2026-03-23,收录于 2026-05-08

Test‐Negative Designs With Multiple Testing Sources

Mengxin Yu, Nicholas P. Jewell

摘要

Test‐negative designs, a form of case‐cohort studies, have been commonly used to assess infectious disease interventions. Early examples of the design included the evaluation of seasonal influenza vaccines in the field. Recently, they have also been widely used to evaluate the efficacy of COVID‐19 vaccines in preventing symptomatic disease for different variants [32]. The design hinges on individuals being tested for the disease of interest; upon recruitment, such individuals are subjected to a definitive test for the presence of the disease of interest (test‐positives) or not (test‐negatives), along with the determination of whether the individual has been exposed to the intervention under study (e.g., vaccination). In most early TND studies, individuals were tested because they were suffering from symptoms consistent with the disease in question, and the TND was a tool to reduce confounding due to healthcare‐seeking behavior. However, in many cases, such as COVID‐19 and Ebola, testing results were available at healthcare facilities for individuals who presented for a variety of reasons in addition to symptoms (e.g., case contact tracing, etc.). Aggregating samples from symptomatic and asymptomatic test results leads to bias in the assessment of the efficacy of the intervention. Here we consider these issues in the context of a specific version of the ‘ multiple reasons for testing problem ,’ motivated by a vaccine trial designed to assess a new Ebola viral disease vaccine (EVD) [22]. Some participants are recruited in the usual TND fashion as they present for care suffering from symptoms consistent with an Ebola diagnosis (and are thus tested); in addition, however, any test‐positive identified in this fashion leads to immediate testing for Ebola for all close contacts of the test‐positive who are likely asymptomatic at that point. We examine a simple approach to estimate the common efficacy of the vaccine intervention based on these two sources of test positives and test negatives, complemented by an assessment of whether efficacy is the same for both sources. The EVD trial was not completed for the fortunate reason that the prevailing disease outbreak ended; nevertheless, the approach here will be important if this trial is ever recommenced or similar trials are conducted in the future.

SiM Causal Inference · 发表于 2026-03-23,收录于 2026-05-01

Powering RCTs for Marginal Effects With GLMs Using Prognostic Score Adjustment

Emilie Højbjerre‐Frandsen, Mark J. van der Laan, Alejandro Schuler

摘要

In randomized clinical trials (RCTs), the accurate estimation of marginal treatment effects is crucial for determining the efficacy of interventions. Enhancing the statistical power of these analyses is a key objective for statisticians. The increasing availability of historical data from registries, prior trials, and health records presents an opportunity to improve trial efficiency. However, many methods for historical borrowing compromise strict type I error rate control. Building on the work by Schuler et al. on prognostic score adjustment for linear models, this paper extends the methodology to the plug‐in analysis proposed by Rosenblum and van der Laan using generalized linear models (GLMs) to further enhance the efficiency of RCT analyses without introducing bias. Specifically, we train a prognostic model on historical control data and incorporate the resulting prognostic scores as covariates in the plug‐in GLM analysis of the trial data. This approach leverages the predictive power of historical data to improve the precision of marginal treatment effect estimates. We demonstrate that this method achieves local semi‐parametric efficiency under the assumption of an additive treatment effect on the link scale. We expand the GLM plug‐in method to include Negative Binomial regression. Additionally, we provide a straightforward formula for conservatively estimating the asymptotic variance, facilitating power calculations that reflect these efficiency gains. Our simulation study supports the theory. Even without an additive treatment effect, we observe increased power or reduced standard error. While population shifts from historical to trial data may dilute benefits, they do not introduce bias. We demonstrate reductions in the estimated variance through an analysis of clinical trial data provided by Novo Nordisk A/S.

SMMR Survival Analysis Clinical Trials · 发表于 2026-03-23

Asymptotic validity of Schoenfeld’s sample size formula for the Cox proportional hazards model via the Wald test approach

Se Yoon Lee

摘要

We revisit the widely used sample size formula for the Cox proportional hazards model, originally proposed by Schoenfeld in 1983. The classical derivation, based on the score test, evaluates the Fisher information under the null hypothesis, overlooking key conditions required for its validity. Using a Wald test framework, we demonstrate that the derivation relies on the risk set proportionality property, wherein the ratio of at-risk counts in the treatment and control arms at observed times matches the randomization ratio. This property typically holds under the null hypothesis or when event rates are low, given a sufficiently large sample size. Our analysis clarifies the asymptotic validity of the formula and shows that violations of this assumption can lead to substantial loss of efficiency, particularly under alternative hypotheses. In contrast, simulation-based approaches remain robust. A retrospective analysis of the ADAURA trial illustrates how simulation-based power analysis could have shortened the study duration compared to the formula-based approach, while still maintaining the type I error rate at the nominal level and preserving the coverage properties of the confidence interval. This work highlights the limitations of the Schoenfeld formula in realistic trial settings and recommends simulation-based methods for planning survival trials, especially when a large treatment effect is expected.

SiM Causal Inference Missing Data & Measurement Error Latent Variables & Mixtures · 发表于 2026-03-23,收录于 2026-05-01

A Latent Variable Approach for Causal Effect Estimation Under Misclassified Treatment Assignment

Yimeng Shang, Yu‐Han Chiu, Lan Kong

摘要

Misclassification in treatment assignment is a common issue in causal inference with observational studies, often leading to biased estimates of causal effects if unaddressed. Several methods have been developed to handle this issue by making use of a validation dataset. This paper proposes a robust latent variable approach for causal effect estimation without the need of validation data. By employing a potential outcome modeling framework that incorporates true treatment assignment as a latent variable, we construct a likelihood function that involves three models: the outcome model, the measurement error model for misclassification, and the propensity score model for treatment assignment. To enhance the robustness against misspecification of the measurement error mechanism, we further incorporate neural networks into the estimation of the measurement error model. The simulation results show that our method performed well under various misclassification assumptions, and that using neural networks reduced the impact of misspecification of functional form for the measurement error model. We illustrate the method using a synthetic dataset derived from the Right Heart Catheterization (RHC) study. This flexible framework mitigates bias and improves the reliability of causal inference when treatment assignment is subject to misclassification and no validation data is available.

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