研究园简易版 · 经典书籍 · 期刊文章 · 有原文 · 完整版(可多选筛选、搜索) · 首页
2026 年的期刊文章 · 第 2 页
共 1756 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2026-11-03 到 2026-10-21。
按期刊、按主题看 →
SiM Clinical Trials · 发表于 2026-11-03,收录于 2026-11-06
Shiyu Feng, Meng Zheng, Yi Zhang, Xuezhi Liang, Minglu Yuan, Wen Chen
摘要
Stepped wedge designs with multiple interventions (M-SWDs) are increasingly used to evaluate complex implementation strategies. In many such trials, participants enter and leave over time, giving rise to open-cohort sampling and complex correlation structures. Existing sample size and power calculation methods are largely restricted to single-intervention designs, closed cohorts, or simplified correlation assumptions. We develop a unified, parameter-based framework for sample size and power calculation in open-cohort M-SWDs. The proposed methodology accommodates several common multi-intervention designs, including concurrent, cross-over, supplementation, and factorial designs. Flexible open-cohort sampling schemes are incorporated to account for participant turnover, and both block-exchangeable and decay correlation structures are considered to capture within-cluster and within-individual dependence over time. Closed-form and generalized variance expressions for intervention effect estimators are derived, enabling efficient power calculation under a wide range of design settings. Extensive simulation studies demonstrate that the proposed formulas provide accurate power estimates, while further power analyses illustrate how the key correlation parameters and participant retention jointly influence power. This work provides practical and theoretically grounded guidance for planning multi-intervention stepped wedge trials under open-cohort sampling, facilitating more realistic and transparent design-stage decision-making.
Biometrika Testing & Uncertainty · 发表于 2026-11-03,收录于 2026-11-06
S Arnold, E Clerico
摘要
We derive explicitly the e-values with optimal (relative) growth rate in the worst case for testing the mean of a bounded random variable, thereby providing the first application of the (RE)GROW quality criteria beyond the assumption of mutually absolutely continuous hypotheses for e-values originally proposed by Grünwald et al. (2024). For both criteria, we explicitly characterise the alternatives that are most difficult to test against and show that they admit a meaningful interpretation. We give two important examples in which REGROW provides a powerful quality criterion to choose optimal e-variables whereas GROW leads to trivial solutions.
SiM Genetics & Omics · 发表于 2026-11-03,收录于 2026-11-06
Weineng Zhu, Xiaoyu Sheng
摘要
High-dimensional pharmacogenomic data pair thousands of molecular features with hundreds of cancer cell lines, complicating conditional density estimation. We study factor-augmented CINDES, which estimates conditional response densities from principal-component scores through binary classification. For an honest pilot/main-sample construction, assuming fresh-score PCA recovery, we derive a risk bound separating conditional-density learning, PCA score error, and representation bias. This guarantee does not cover the implemented procedure, which reuses observations for PCA and CINDES. In favorable strong-factor simulations, PCA improved a fixed CINDES pipeline, but PCA-MDN achieved lower negative log-likelihood (NLL) and integrated L 1 error in the matched comparison. Stress tests showed losses from omitted response-relevant low-variance directions and weak factor recovery, with sample-size-dependent factor-count effects. Across all 692 eligible source-specific genomics of drug sensitivity in cancer (GDSC) units and five splits, mean held-out NLL was 1.5414 for PCA-CINDES, versus 1.5661 and 1.5557 for ElasticNet and PCA/Ridge Gaussian models; PCA-MDN was better at 1.5156. A previously fixed 35-unit audit favored PCA-CINDES over Raw-X CINDES (mean paired NLL difference - 0 . 178 ) and found lower measured computational cost. The evidence supports PCA compression when retained factors capture response information and are accurately estimated; it does not establish robustness to general nonlinear or nonfactor structures, or CINDES as the preferred second-stage estimator. The application concerns preclinical prediction rather than clinical utility.
Biometrics Clinical Trials Longitudinal Data · 发表于 2026-10-31,收录于 2026-11-06
Chunyi Zhang, Ying Yuan, Ruitao Lin, Suyu Liu
摘要
Decentralized clinical trials (DCTs) offer significant advantages in enhancing trial accessibility, reducing patient burden, and improving population representativeness. These benefits are particularly pronounced in longitudinal trials, where patient outcomes are measured repeatedly over time. However, existing literature lacks appropriate sample size formulas tailored to longitudinal DCTs, which must account for heterogeneous measurement sources and increased uncertainty introduced by decentralized data collection. We propose a unified sample size calculation method for randomized longitudinal DCTs under the generalized linear model framework. The proposed formula is available in closed form and flexibly accommodates a variety of data collection designs, including the simple hybrid design, where all patients contribute both onsite and offsite data, and the stratified hybrid design, where different patient groups may undergo distinct data collection protocols. We also develop estimation procedures to estimate the unknown parameters and compute the sample statistics. Our method further generalizes to accommodate heterogeneous intraclass correlation coefficients, unequal variances between treatment arms, and varying numbers of data collection types or strata. Extensive simulation studies show that the proposed sample size formula achieves robust control of the type I error rate and maintains the desired power across a wide range of realistic scenarios.
SiM Longitudinal Data Meta-Analysis & Data Integration Bayesian Methods · 发表于 2026-10-30,收录于 2026-11-06
Md Tuhin Sheikh
摘要
Incorporating external data can improve precision in longitudinal studies, but effective borrowing requires that information from different studies be comparable. Within a Bayesian framework, power priors are widely used for external data integration through likelihood discounting. However, they assume that parameters are on a common information scale. In longitudinal settings, differences in follow-up duration, measurement variability, and correlation structure, along with random effects, can lead to differences in effective Fisher information across studies. This complicates parameter comparability and poses challenges for existing methods, which can result in biased or inefficient inference. We propose a scale-transformed power prior for external data integration in longitudinal studies. The method is developed within a linear mixed-effects modeling framework and rescales historical parameters to the current study scale using Fisher information before borrowing. Conditional Fisher information accounts for unobserved random effects and adjusts for study-specific variability. To address differences in variability and correlation structure, we implement a partial borrowing strategy that applies borrowing to fixed effects while allowing variance components to remain study-specific. We introduce a generalized formulation with an additive adjustment component to allow for remaining differences after scaling. Theoretical results characterize how differences in information scale lead to bias under conventional power priors and establish conditions under which the proposed method improves bias and precision. Simulation studies demonstrate improved performance and provide guidance on settings where scale transformation is beneficial. An application to prostate cancer clinical trial data demonstrates the importance of the proposed method.
Biometrics High-Dimensional · 发表于 2026-10-30,收录于 2026-11-06
Sung Hee Park, Xin Zhang, Elizabeth H Slate, Shuchun Sun, Hai Yao
摘要
Sexual dimorphism is a critical factor in many biological and medical research fields. In biomechanics and bioengineering, understanding sex differences is crucial for studying musculoskeletal conditions such as temporomandibular disorder. This paper focuses on the association between the craniofacial skeletal morphology and temporomandibular joint related masticatory muscle attachments to discern sex differences. Data were collected from 11 female and 10 male cadaver heads to investigate sex-specific relationships between the skull and muscles. We propose a conditional cross-covariance reduction (CCR) model, designed to examine the dynamic association between two sets of random variables conditioned on a third binary variable (e.g., sex), highlighting the most distinctive sex-related relationships between skull and muscle attachments in the human cadaver data. Under the CCR model, we employ a sparse singular value decomposition algorithm. We further introduce a sequential permutation for selecting sparsity method to identify important variables and determine the optimal number of selected variables.
Biometrika Nonparametric & Semiparametric · 发表于 2026-10-29
Z Xu, D Witten, A Shojaie
摘要
Summary: We consider statistical inference under a semi-supervised setting where we have access to both labelled and unlabelled observations. We ask the question: under what circumstances, and by how much, can incorporating the unlabelled data improve upon inference using the labelled data? To answer this question, we investigate semi-supervised inference through the lens of semiparametric efficiency theory. For general pathwise differentiable targets under semiparametric models, we derive the efficiency lower bound for asymptotically linear semi-supervised estimators. Perhaps not surprisingly, this bound coincides with the results in classical missing data theory when ‘positivity’ holds, i.e., when the limiting proportion of labelled observations is strictly positive. However, unlike missing data theory, our theory can handle the boundary regime in which the proportion of labelled observations converges to zero, which is in fact the setting of primary interest in the context of semi-supervised learning. We further characterize a broad class of well-specified targets for which unlabelled data cannot yield efficiency gains. These results shed insight into the promise and limitations of the very active field of prediction-powered inference. We propose an efficient semi-supervised estimator that attains the efficiency bound, and a prediction-powered estimator that incorporates black-box predictions and is guaranteed to be no worse than the supervised baseline in asymptotic variance. We illustrate our theoretical results via simulation studies and a real-data application.
SiM Latent Variables & Mixtures Missing Data & Measurement Error · 发表于 2026-10-29,收录于 2026-11-06
Lingyi Hu, Yao Cui, Qi Yang, Shan Tang, Shanjun Mao
摘要
Predicting postoperative complications is a critical task for improving patient care and optimizing resource management. We formulate the task as a binary classification problem based on two‐stage clinical data: preoperative and intraoperative variables. A key difficulty is that intraoperative data are systematically missing for preoperative prediction. Existing methods, which typically rely solely on preoperative data or single intraoperative indicator imputation, generally fail to capture important latent intraoperative information. To provide a more robust solution, we propose P4MC (Preoperative, Partial intraoPerative, and Postoperative Monte Carlo classification). This approach explicitly models the dependence between the two stages to recover latent intraoperative information. Specifically, P4MC first applies partial sufficient dimension reduction (PSDR) to compress intraoperative variables into a low‐dimensional sufficient reduction. We then estimate the conditional distribution of this reduction given preoperative variables using nonparametric density estimation and generate latent intraoperative features for classification via Monte Carlo sampling. A simulation study demonstrates the improved predictive performance and stability of P4MC when combined with standard machine learning methods. Applications to two clinical cohorts, laparoscopic pancreaticoduodenectomy and hepatocellular carcinoma, further show that P4MC outperforms conventional approaches and yields clinically interpretable insights. An open‐source R program is provided in the P4MC package, available at https://github.com/501fox/P4MC.git .
JASA Spatial & Time Series · 发表于 2026-10-29
Haotian Xu, Carlos Misael Madrid Padilla, Oscar Hernan Madrid Padilla, Daren Wang
摘要
In this work, we propose new matrix- and tensor-based methodologies for estimating multivariate intensity functions of inhomogeneous point processes. By viewing multivariate intensity functions as infinite-dimensional matrices or tensors within function spaces, our algorithms attain the optimal bias-variance trade-off, yielding rate-optimal estimation error, with model complexity governed by matrix or tensor ranks. They substantially improve estimation accuracy, while simultaneously reducing computational cost. To illustrate the adaptivity of the proposed framework, we show that many fundamental classes of multivariate functions, including additive and mean-field models, admit finite-rank tensor representations. We apply our method to a four-dimensional U.S. Geological Survey earthquake dataset, comprising features such as latitude, longitude, depth, and magnitude. Our tensor estimator recovers localized seismicity patterns (California, Oklahoma, Pacific Northwest, north-central U.S.), whereas the kernel baseline oversmooths them.
SiM Survival Analysis Nonparametric & Semiparametric · 发表于 2026-10-29,收录于 2026-11-06
Daniel J Sharpe, Georgia Yates, Tuli De, Renuka Kandikatla, Prianka Singh, Matthew Dixon, Kiran Davé, Jackie Vanderpuye-Orgle, Murat Kurt
摘要
OBJECTIVES: Overall survival (OS) extrapolations from mixture cure models (MCMs) are often sensitive to the cure fraction, which is challenging to estimate when data are immature. Furthermore, in some cancer populations, it may be unreasonable to assume curative effects among patients who are not event-free. We addressed these problems using bivariate copula MCMs to jointly model OS and progression- (PFS) or recurrence-free survival (RFS), with a single event-free cure fraction parameter. METHODS: We proposed a practical framework for parametric survival copula model selection and interpretation, which involves estimating five representative candidate copula functions and examining OS distributions conditioned on intermediate event times. Bivariate MCMs with event-free cure fraction, based on previously selected marginal distributions, were estimated for two studies of nivolumab with contrasting survival patterns, namely, in previously treated advanced esophageal squamous cell carcinoma (OS-PFS correlation) and resected melanoma (OS-RFS correlation). RESULTS: OS extrapolations from bivariate MCMs with an effective PFS (or RFS) cure fraction were more conservative than from univariate MCMs. Although patient-level OS-PFS/RFS correlation coefficients were sensitive to the copula, model averaging and inspecting estimated conditional OS distributions enabled clinically meaningful inference on the prognostic impact of intermediate events. CONCLUSIONS: Bivariate MCMs with event-free cured subpopulation incorporated curative effects into OS extrapolations more cautiously than conventional MCMs, based on a more rapidly maturing composite endpoint. The model is especially useful when OS data are immature and if post-event cure is clinically implausible. Employing alternative copulas that cover archetypal association patterns mitigated model selection bias in patient-level correlation analysis.
SiM Causal Inference · 发表于 2026-10-28,收录于 2026-10-29
Xiaoya Wang, Richard J Cook, Yeying Zhu, Tugba Akkaya-Hocagil, R Colin Carter, Sandra W Jacobson, Joseph L Jacobson, Louise M Ryan
摘要
Methods for causal inference are well developed for binary and continuous exposures, but in many settings, the exposure has a substantial mass at zero-such exposures are called semi-continuous. We propose a general causal framework for such semi-continuous exposures, together with a two-stage estimation strategy. A two-part propensity structure is introduced for the semi-continuous exposure, with one component for exposure status (exposed vs unexposed) and another for the exposure level among those exposed, and incorporates both into a marginal structural model that disentangles the effects of exposure status and dose. The two-stage procedure sequentially targets the causal dose-response among exposed individuals and the causal effect of exposure status at a reference dose, allowing flexibility in the choice of propensity score methods in the second stage. We establish consistency and asymptotic normality for the resulting estimators, and characterize their limiting values under misspecification of the propensity score models. Simulation studies evaluate finite sample performance and robustness, and an application to a study of prenatal alcohol exposure and child cognition demonstrates how the proposed methods can be used to address a range of scientific questions about both exposure status and exposure intensity.
SiM Survival Analysis High-Dimensional · 发表于 2026-10-28,收录于 2026-10-29
Li Jin, Tao Hu, Liming Xiang
摘要
The growing availability of large-scale survival data streams present significant challenges for statistical inference particularly in the presence of a cure fraction, which complicates estimation and increases computational burden. To address these challenges, we propose an online updating mixture cure model framework. Our approach employs a batchwise renewable estimation procedure that aggregates historical information into summary statistics to update the model without retaining historical individual-level data, and thus eliminating the need for full data storage. A renewable penalized expectation-maximization algorithm is developed to enable simultaneous parameter estimation and variable selection. Furthermore, a tuning-free strategy is incorporated to enhance computational efficiency. The asymptotic properties of the proposed method are established. Simulation studies and an application to breast cancer data from the Surveillance, Epidemiology, and End Results database demonstrate the method's estimation accuracy, predictive performance, and computational scalability.
SiM Meta-Analysis & Data Integration · 发表于 2026-10-28,收录于 2026-10-29
Muhammad Mohsin, Stephan Morgenthaler
摘要
The DerSimonian and Laird (DL) method, commonly used in meta-analysis, often underestimates the between-study variance. The underestimation causes the confidence intervals to have less than the expected 95% coverage, resulting in inconclusive research findings. The present study introduces the use of the jackknife technique to improve the DL between-study variance estimator for meta-analyses of proportions from single-arm studies, as well as suggests certain transformations in combination with the jackknife technique. The proposed methodology is intended for settings in which the parameter of interest is an event probability or proportion, particularly for small numbers of studies with large variation. Simulation studies for binomial observations are conducted to compare the relative biases and coverage probabilities of the mean ( μ ) obtained by the DL, Paule and Mandel (PM), Restricted Maximum Likelihood (REML), and Hartung-Knapp-Sidik-Jonkman (HKSJ) methods with our proposed DerSimonian and Laird jackknife (DLJ) and transformed DerSimonian and Laird jackknife (TDLJs) methods. The simulation results show that our proposed DLJ and TDLJ estimators better quantify the between-study variance in meta-analysis of binary events, especially for small numbers of studies with larger variation and high probability of μ respectively. Three real life examples are presented to determine the performance of our new proposed between-study variance estimators.
SMMR Clinical Trials Longitudinal Data · 发表于 2026-10-28,收录于 2026-10-29
Hao Wang, Guangyu Tong, Heather Allore, Kelsey L Grantham, Monica Taljaard, Fan Li
摘要
In stepped wedge cluster randomized trials (SW-CRTs), the intervention is rolled out to clusters sequentially over multiple periods. It is common practice to analyze data from SW-CRTs using linear mixed models that treat time as discrete. However, a recent systematic review found that 95.1% of cross-sectional SW-CRTs recruit individuals continuously over time. Despite the prevalence of continuous recruitment, there has been limited guidance on how to draw model-robust inference for such designs. In this article, we conduct simulation studies to examine the performance of discrete-time linear mixed models in the case of continuous recruitment designs with a continuous outcome. Specifically, we generate outcomes under a continuous-time exponential decay correlation structure with either discrete or continuous fixed period effects. We consider constant and exposure-time-dependent intervention effects, with and without random intervention effects across clusters. We then analyze the simulated data using linear mixed models under three discrete-time working correlation structures: simple exchangeable, nested exchangeable, and discrete-time exponential decay, with robust sandwich variance estimation. Our results demonstrate that discrete-time analysis often yields negligible bias and that the robust variance estimator with the Mancl and DeRouen correction consistently achieves nominal coverage and type I error rate. However, when recruitment patterns vary systematically between control and intervention periods, discrete-time analyses give slightly biased intervention effect estimates. Finally, we illustrate these findings by reanalyzing a completed SW-CRT.
Biometrika Other · 发表于 2026-10-27,收录于 2026-10-29
S Y Li, D A van Dyk, M Autenrieth
摘要
This paper presents a novel method for analytical derivations of marginal densities using the fractional derivatives of moment-generating functions. Although the method requires likelihood functions to take specific forms, its assumptions are otherwise modest. It only requires that the prior moment-generating functions exist, are finite, and are continuous and differentiable at certain points. We also present the probabilistic and statistical insights behind this method.
Biometrika Optimization & Learning Theory · 发表于 2026-10-27,收录于 2026-10-29
Haoshu Xu, Hongzhe Li
摘要
We propose a novel test for assessing partial effects in Fréchet regression on the Bures-Wasserstein manifold. Under the null hypothesis, we show that the statistic admits a degenerate V-statistic approximation whose limiting distribution is aweighted mixture of chi-squared random variables, with weights determined by the eigenvalues of an integral operator associated with a reproducing kernel Hilbert space (RKHS) kernel. We establish the asymptotic validity and consistency of the proposed test. Its finite-sample performance is examined through simulation studies.We apply the proposed test to study the effect of age, while controlling for other covariates, on gene co-expression structure in single-cell data.
JASA Causal Inference · 发表于 2026-10-27,收录于 2026-10-29
Soumyabrata Kundu, Peng Ding, Jingshu Wang, Xinran Li
摘要
The test-negative design has become popular for evaluating the effectiveness of post-licensure vaccines using observational data. In addition to its logistical convenience on data collection, the design is also believed to control for the differential health-care-seeking behavior between vaccinated and unvaccinated individuals, an important while often unmeasured confounder between the vaccination and infection. Hence, the design has been employed routinely to monitor seasonal flu vaccines and more recently to measure the COVID-19 vaccine effectiveness. Despite its popularity, the design has been questioned, in particular about its ability to fully control for the unmeasured confounding. In this paper, we explore deviations from a perfect test-negative design, and propose various sensitivity analysis methods for estimating the effect of vaccination measured by the causal odds ratio on the subpopulation of individuals with good health-care-seeking behavior. We start with point identification of the causal odds ratio under a test-negative design, comparing different forms of identification assumptions and their corresponding estimands. We then propose two approaches for conducting sensitivity analysis, addressing the influence of the unmeasured confounding in two different ways. Specifically, one approach investigates partial control for unmeasured confounding in the test-negative design, while the other examines the impact of unmeasured confounding on both vaccination and infection. Furthermore, we combine these approaches to provide narrower bounds on the true causal odds ratio, and further sharpen the bounds by restricting the treatment effect heterogeneity. Finally, we apply the proposed methods to evaluate the effectiveness of COVID-19 vaccines using observational data from test-negative designs.
JASA Causal Inference Optimization & Learning Theory · 发表于 2026-10-27,收录于 2026-10-29
Ruijia Zhang, Xiangyu Zhang, Zhengling Qi, Yue Wu, Yanxun Xu
摘要
Dynamic treatment regimes (DTRs) provide a principled framework for optimizing sequential decision-making in domains where decisions must adapt over time in response to individual trajectories, such as healthcare, education, and digital interventions. However, existing statistical methods often rely on strong positivity assumptions and lack robustness under partial data coverage, while offline reinforcement learning approaches typically focus on average training performance, lack statistical guarantees, and require solving complex optimization problems. To address these challenges, we propose POLAR, a novel pessimistic model-based policy learning algorithm for offline DTR optimization. POLAR estimates the transition dynamics from offline data and quantifies uncertainty for each history-action pair. A pessimistic penalty is then incorporated into the reward function to discourage actions with high uncertainty. Unlike many existing methods that focus on average training performance or provide guarantees only for an oracle policy, POLAR directly targets the suboptimality of the final learned policy and offers theoretical guarantees, without relying on computationally intensive minimax or constrained optimization procedures. To the best of our knowledge, POLAR is the first model-based DTR method to provide both statistical and computational guarantees, including finite-sample bounds on policy suboptimality. Empirical results on both synthetic data and the MIMIC-III dataset demonstrate that POLAR outperforms state-of-the-art methods and yields near-optimal, history-aware treatment strategies. *(abstract via OpenAlex)*
Biometrics High-Dimensional · 发表于 2026-10-27,收录于 2026-10-29
Junfeng Cui, Junhui Wang, Changliang Zou, Yan Zhou
摘要
Existing changepoint localization methods for high-dimensional regression models typically focus on scenarios where the regression coefficients follow a piecewise constant structure over time. However, this assumption can be overly restrictive in many practical applications due to the complex temporal dynamics of the coefficients. Additionally, in sparse regression models, the primary interest sometimes lies not in changes in the magnitudes of the coefficients, but rather in identifying the time points at which covariates transition between inactive and active states. In this paper, we propose a simple yet effective method for detecting changepoints in the support set, allowing the coefficients to vary continuously over time. Theoretically, we establish the minimax lower bound on the localization error that nearly matches the upper bound achieved by the proposed method, and show that the required signal-to-noise condition is essentially the weakest possible based on the impossibility region in a minimax sense. Extensive numerical studies further demonstrate the superior performance of the proposed method. It is also applied in the analysis of real air quality data from Northern Taiwan in 2015. An R package "SCP" is developed to implement the proposed method.
Biometrika Other · 发表于 2026-10-27,收录于 2026-10-29
Seongchan Lee, Suman Cha, Ilmun Kim
摘要
We study the problem of conditional two-sample testing, which aims to determine whether two populations have the same distribution after accounting for confounding factors. This problem commonly arises in various applications, such as domain adaptation and algorithmic fairness, where comparing two groups is essential while controlling for confounding variables. We begin by establishing a hardness result for conditional two-sample testing, demonstrating that no valid test can have significant power against any single alternative without proper assumptions. We then introduce two general frameworks that implicitly or explicitly target specific classes of distributions for their validity and power. Our first framework allows us to convert any conditional independence test into a conditional two-sample test in a black-box manner, while preserving the asymptotic properties of the original conditional independence test. The second framework transforms the problem into comparing marginal distributions with estimated density ratios, which enables direct application of existing methods for marginal two-sample testing. We demonstrate this idea in a concrete manner with classification and kernel-based methods. Finally, simulation studies are conducted to illustrate the proposed frameworks in finite-sample scenarios.
Biometrics Missing Data & Measurement Error Bayesian Methods Functional Data · 发表于 2026-10-27,收录于 2026-10-29
Ziren Jiang, Lei Xuan, Eric F Lock, Erjia Cui
摘要
Missing data are pervasive in modern functional datasets, where trajectories are often sparsely or irregularly observed. Although functional principal component analysis is widely used to reconstruct incomplete curves, existing approaches typically employ single imputation, leading to overly optimistic inferences in downstream analyses. To address these challenges, we develop a novel Bayesian multiple imputation framework for functional data (BAMIFun). For single-level functional data, we impose a Bayesian low-rank model that incorporates penalized spline representations to enforce smoothness of the functional domain and derive an efficient Gibbs sampler algorithm for posterior computation. In addition, we demonstrate and validate how to properly account for estimation uncertainties in downstream analysis. Furthermore, we extend the framework to multiway functional data using Functional Tensor Singular Value Decomposition model, enabling Bayesian multiple imputation in settings not supported by existing methods. Simulation studies show that BAMIFun achieves substantially improved coverage and more reliable downstream inference compared to existing methods, while maintaining similar imputation accuracy. Case studies using a physical activity dataset and an infant gut microbiome dataset further demonstrate the practical advantages of our proposed methods under severe missingness.
JASA Causal Inference Genetics & Omics · 发表于 2026-10-27,收录于 2026-10-29
Haoyu Yang, Ruoyu Wang, Yuzhou Lin, Xihong Lin
摘要
Large-scale mediation testing has attracted considerable attention in epigenome-wide studies, intending to determine whether the effect of an exposure on an outcome is mediated through specific DNA methylation sites. Statistical inference for the mediation effects is challenged by the complexity of testing a large number of composite null hypotheses. We propose a tail likelihood ratio (TLR) test that is robust to variations in the proportion of null cases, signal strength, and signal distribution, achieving higher power for marginal mediation testing than existing methods while maintaining robust error rate control across diverse scenarios, with false discovery rate control established theoretically under independence and supported by simulation under dependence. The proposed TLR method features a powerful test statistic whose distribution under the composite null is derived analytically. Meanwhile, since methylations are often correlated, we account for the correlation of dependent test statistics using a principal component-based approximation. Simulations demonstrate that the proposed method maintains accurate error rate control in multiple comparisons and exhibits power gains compared to existing methods. We apply the proposed method to the Normative Aging Epigenetic Study and identify 27 of the 484,548 CpG sites of DNA methylation measured on 632 men that mediate the effect of smoking on lung function. *(abstract via OpenAlex)*
Biometrika Bayesian Methods · 发表于 2026-10-26
Quan Zhou
摘要
We study the importance-tempered random-walk Metropolis (RWM) algorithm on Rd, which targets the tempered density πβ(x)∝(π(x))β with β∈(0,1), and corrects for the resulting bias by importance weighting. The dynamics of this sampler can be described by a continuous-time Markov chain whose embedded chain is the tempered random-walk Metropolis. We prove that when the tails of π decay at rate ‖x‖−γ, this continuous-time chain is uniformly ergodic if and only if dγ<β<(γ−2)γ. Furthermore, the global spectral gap of this chain can be controlled via the spectral gap of the chain restricted to a ball with radius growing polynomially in d. Our theory and and simulation studies show that, with a suitable choice of β, importance tempering can substantially reduce sensitivity to initialization and significantly improve efficiency over the standard random-walk Metropolis.
SMMR Survival Analysis · 发表于 2026-10-26,收录于 2026-10-29
Giuseppe Loffredo, Elvira Romano, Fabrizio Maturo
摘要
Functional survival models are key tools for analyzing time-to-event data with complex predictors, such as functional or high-dimensional inputs. Despite their predictive strength, these models often lack interpretability, which limits their value in practical decision-making and risk analysis. This study investigates two key survival models: the functional survival tree (FST) and the functional random survival forests (FRSFs). It introduces novel methods and tools to enhance the interpretability of FST models and improve the explainability of FRSF ensembles. Applications to simulated data and to a real intensive care dataset illustrate that the proposed approaches provide post-hoc graphical and functional summaries that facilitate the interpretation of FSTs and the explanation of forest-level predictions.
SiM Bayesian Methods Causal Inference Clinical Trials · 发表于 2026-10-26,收录于 2026-10-29
Ruyi Liu, Joshua L Warren, Yuki Ohnishi, Donna Spiegelman, Liangyuan Hu, Fan Li
摘要
In cluster-randomized trials (CRTs), entire clusters of individuals are randomized to treatment, and outcomes within a cluster are typically correlated. While frequentist approaches are standard practice for CRT analysis, Bayesian methods have emerged as a strong alternative. Previous work has investigated the use of Bayesian hierarchical models for continuous, binary, and count outcomes in CRTs, but these approaches focus on model-based treatment effect coefficients as the target estimands, which may have ambiguous interpretation under model misspecification and informative cluster size. In this article, we introduce a calibrated Bayesian procedure for estimand-aligned analysis of CRTs even in the presence of potentially misspecified models. We propose estimators targeting both the cluster-average treatment effect (cluster-ATE) and individual-average treatment effect (individual-ATE), particularly in scenarios with informative cluster sizes. We additionally explore strategies for summarizing the posterior samples that can achieve the frequentist coverage even under working model misspecification. We provide simulation evidence to demonstrate the model-robustness property of the proposed estimators in CRTs, and further investigate the impact of covariate adjustment as well as the use of more flexible Bayesian nonparametric working models in the CRT context.
SMMR Optimization & Learning Theory · 发表于 2026-10-25,收录于 2026-10-26
Kentaro Takeda, Masahiro Kojima
摘要
The emergence of molecular-targeted agents and immune-oncology therapies has fundamentally transformed oncology drug development, necessitating evolution beyond traditional dose-finding approaches designed for cytotoxic agents. While conventional agents exhibit predictable monotonic dose-response relationships, novel anticancer agents often demonstrate plateau-effect patterns where higher doses may compromise therapeutic benefit, requiring identification of optimal biological doses that balance efficacy and tolerability. The FDA's Project Optimus initiative emphasizes comprehensive dose optimization through parallel randomized cohorts and patient backfilling to better understand pharmacological profiles across multiple dose levels. Contemporary drug development increasingly prioritizes combination therapy alongside monotherapy evaluation, yet existing designs typically assume equivalent roles for both agents, diverging from clinical practice where novel agents combine with established treatments having limited dose options. This article proposes a seamless dose-optimization design that adaptively evaluates both monotherapy and combination therapy based on efficacy and toxicity outcomes through adaptive subtrials with patient backfilling capabilities. The model-assisted framework employs predetermined Bayesian optimal boundaries, eliminating real-time model fitting while accommodating evaluation of both monotherapy and combination therapy and enabling sequential enrollment with strategic backfilling. Simulation studies demonstrate robust performance across diverse dose-response patterns relevant to contemporary oncology. The design addresses critical gaps between methodological assumptions and clinical reality, offering a practical approach that integrates monotherapy and combination therapy evaluation with efficacy-toxicity-based backfilling. This framework ensures straightforward implementation while maintaining flexibility for complex dose-response relationships characteristic of novel anticancer agents, providing significant advantages for clinical implementation in the era of targeted and immune-oncology therapeutics.
Biometrika Nonparametric & Semiparametric · 发表于 2026-10-23,收录于 2026-10-26
F Camirand Lemyre, A Delaigle, J Mao
摘要
We consider nonparametric estimation of the nonstationary density of streaming data collected continuously over time. Those types of data are often available only for a limited time after being observed, for example due to storage or data privacy constraints. Therefore, analyzing them requires dynamic approaches that do not require repeated access to past data. A suitable iterative kernel estimator of nonstationary densities has been developed by Rutkowski (1989); it requires choosing important tuning parameters each time a new observation arises, for which no adequate practical selection criterion has been developed yet.We propose a data-driven dynamic selection of those parameters, which can be implemented iteratively and requires only sequential access to consecutive blocks of the most recent data. To choose the sizes of the blocks, we propose a theoretically founded approach based on properties of cross-validation criteria. We illustrate the broader potential of our ideas by suggesting a dynamic blockwise practical implementation of another kernel estimator of nonstationary densities developed by Hall et al. (2006).We apply our technique to simulated data and to light curve data from the NASA Kepler mission.
JASA Latent Variables & Mixtures · 发表于 2026-10-23,收录于 2026-10-26
Yanxin Jin, Peiyao Cai, Janice Chen, Yang Ning, Kean Ming Tan
摘要
In many scientific fields such as genomics and neuroscience, data often exhibit complex structures where features can belong to multiple clusters simultaneously and vary over time. Traditional clustering methods such as k-means and hierarchical clustering typically assume static cluster structures and perform hard assignments, limiting their ability to capture the underlying overlapping and time-varying cluster structures. To address these challenges, we propose a novel time-varying Auto-Align clustering (AAclust) framework that effectively estimates and tracks dynamic changes in overlapping cluster memberships over time. Specifically, we employ a time-varying factor model, in which the overlapping cluster structures are modeled through a factor loading matrix that is allowed to vary over time. Theoretically, we show that the population factor loading matrix is identifiable up to a signed permutation across time, i.e., the population overlapping cluster memberships are identifiable. Furthermore, we show that our framework can consistently estimate the number of clusters and the overlapping cluster memberships for the features. By leveraging cross-covariance matrices between any two time points, our proposed method automatically aligns cluster labels over time without requiring post-hoc label matching based on similarity measures like the Rand index. We evaluate the performance of our proposed method through numerical experiments and an application to a functional magnetic resonance imaging dataset under continuous natural stimuli. *(abstract via OpenAlex)*
SiM Causal Inference Clinical Trials · 发表于 2026-10-22,收录于 2026-10-26
Yongdong Ouyang, Monica Taljaard, James P Hughes, Fan Li
摘要
Stepped‐wedge cluster randomized trials (SW‐CRTs) evaluate interventions that are sequentially rolled out across clusters. Standard analyses of SW‐CRTs typically use immediate‐treatment (IT) models, which assume the treatment effects begin at crossover and remain constant thereafter. When treatment effects vary with exposure duration, IT models may misrepresent the target effects. Exposure‐time indicator (ETI) models address this by allowing treatment effects to differ by time since exposure and can properly target the time‐averaged treatment effect (TATE) and long‐term effect (LTE). Like IT models, ETI models require specification of a random‐effects structure, which may be misspecified. While robust variance estimators (RVEs) theoretically protect against such misspecification, their empirical performance under ETI models is not well understood. We review RVEs for ETI models and evaluate them in comprehensive simulation studies with continuous and binary outcomes under correctly specified (binary only) and misspecified random‐effects structures. We compare the classic sandwich variance, Kauermann–Carroll (KC), Mancl–DeRouen (MD), and Morel–Bokossa–Neerchal (MBN) bias‐corrected sandwich variance estimators for inference on the TATE and LTE in primary simulations and additionally evaluate Fay–Graubard (FG) in a sub‐study. Our simulations show that under misspecified random‐effects structures, model‐based standard errors (SE) produced undercoverage, whereas RVEs improved performance under the ETI models. For continuous outcomes, MD with a t ‐distribution and degrees of freedom equal to the number of clusters minus two gave the closest to nominal coverage probabilities. Surprisingly, however, MBN was the only consistently reliable option for binary outcomes. MD, however, could be unstable in one‐cluster‐per‐sequence designs because of data sparsity. Across scenarios, both model‐based SE and RVE for LTE were unstable, indicating that greater caution is needed when targeting LTE under ETI models.
Biometrika Other · 发表于 2026-10-22,收录于 2026-10-26
Xinbing Kong, Liwen Zhang, Tong Zhang
摘要
In this paper, we introduce a generalized likelihood ratio test for assessing whether the factor loading matrix in high-dimensional data exhibits a Kronecker structure, thereby offering a rigorous tool for validating two-way matrix factor models. A central challenge in likelihood-based inference for factor models arises from inherent rotational non-identifiability, which prevents the likelihood function from admitting a standard second-order Taylor expansion. To address this issue, we augment the quasi-likelihood with a quadratic identification penalty that removes rotational degeneracy locally. Building on this construction, we derive the asymptotic distribution of the proposed statistic under the null hypothesis and unveil a new Wilks-type phenomenon in a high-dimensional setting. Under the alternative hypothesis, we further establish the consistency of the test, demonstrating its ability to detect deviations from the Kronecker structure. Simulation studies demonstrate the finite-sample effectiveness of the proposed method. Furthermore, applications to real datasets illustrate the practical utility of the proposed procedure by assessing the presence of Kronecker structure in factor loading matrices.
SiM Other · 发表于 2026-10-22,收录于 2026-10-26
Anyin Feng, Xiangbin Hu, Wen Su, Xingqiu Zhao
摘要
This article investigates a partially linear additive reversed mean model for panel count data with an informative terminal event. By anchoring the time origin at the terminal event, the proposed model characterizes the recurrent event process prior to the occurrence of the terminal event. The partially linear additive form provides a flexible framework for modeling covariate effects. The B-spline sieve approximation is employed for both the baseline reversed mean function and the link function. To make the optimization of the loss function feasible, we treat the distribution function of the terminal event as a nuisance functional parameter and develop a two-stage estimation procedure. We establish the asymptotic properties of the proposed estimator, including consistency, convergence rate, and asymptotic distribution. Simulation studies demonstrate that the proposed estimator performs well in finite-sample settings. The proposed method is further illustrated through an application to the Chinese longitudinal healthy longevity survey (CLHLS).
SiM Survival Analysis Nonparametric & Semiparametric · 发表于 2026-10-22,收录于 2026-10-26
Xinyi Wang, Peijie Wang, Xinyuan Song
摘要
Motivated by the National Alzheimer's Coordinating Center (NACC) study, this paper addresses the marginal analysis of clustered data under a complex setting involving interval censoring, competing risks, and informative cluster size, for which existing methods are not readily applicable. We consider a class of marginal generalized odds-rate transformation models for the cumulative incidence function in the presence of interval-censored competing risks, with the Fine-Gray model and the sub-distribution proportional odds model included as special cases. For statistical inference, a cluster-weighted sieve maximum pseudo-likelihood estimation procedure is developed, in which B-splines are used to approximate the nonparametric components of the semiparametric models. This method effectively reduces the original infinite-dimensional estimation problem to a finite-dimensional one. We establish the asymptotic properties of the resulting parameter estimators and conduct extensive simulation studies to evaluate their finite-sample performance. The proposed methodology is applied to the motivating NACC data to investigate potential risk factors for the cumulative incidence of Lewy body dementia.
SiM Bayesian Methods Clinical Trials · 发表于 2026-10-22,收录于 2026-10-26
David Jesse, Diane Uschner, Markus C Elze, Tim Friede, Francois Mercier
摘要
The idea of leveraging historical control data in randomized clinical trials (RCTs) has gained a lot of attention, in particular for rare disease and pediatric indications where conducting traditional and fully powered RCTs can be challenging. In this regard, Bayesian dynamic borrowing (BDB) methods, which adapt the amount of borrowing based on the outcome heterogeneity between the current and historical data, are well established. In recent years, an increasing number of related methods have been published that combine the BDB framework with approaches that account for baseline covariates. The underlying rationale is that some of the heterogeneity may be explained systematically by variations in the distributions of these characteristics. While many such methods are now available, direct comparisons of them are scarce and there is no consensus on which one can be considered the state of the art. We address this issue by conducting a neutral comparison of three methods of this class, namely the propensity score weighted commensurate prior, propensity score weighted multi-source exchangeability models, and Bayesian additive regression trees. We assess their performance for continuous outcomes through extensive simulations and further evaluate them using two examples based on real RCT data. Our study reveals potential issues with some of these approaches and provides recommendations for choosing a particular method for real RCTs.
SiM Bayesian Methods · 发表于 2026-10-22,收录于 2026-10-26
Yinjun Zhao, Yuanjia Wang, Ying Liu
摘要
Understanding the biological and behavioral heterogeneity underlying psychiatric disorders is critical for advancing precision diagnosis, treatment, and prevention. This paper addresses the scientific question of how multimodal data, which spans clinical, cognitive, and neuroimaging measures, can be integrated to identify biologically meaningful subtypes of mental disorders. We introduce Mixed INtegrative Data Subtyping (MINDS), a Bayesian hierarchical model designed to jointly analyze mixed-type data for simultaneous dimension reduction and clustering. Using data from the adolescent brain cognitive development (ABCD) Study, MINDS integrates clinical symptoms, cognitive performance, and brain structure measures to subtype attention-deficit/hyperactivity disorder (ADHD) and obsessive-compulsive disorder (OCD). Our method leverages Pólya-Gamma augmentation for computational efficiency and robust inference. Simulations demonstrate improved stability and accuracy compared to existing clustering approaches. Application to the ABCD data reveals clinically interpretable subtypes of ADHD and OCD with distinct cognitive and neurodevelopmental profiles. These findings show how integrative multimodal modeling can enhance the reproducibility and clinical relevance of psychiatric subtyping, funding data-driven policies for early identification and targeted interventions in mental health.
SiM Missing Data & Measurement Error Clinical Trials · 发表于 2026-10-22,收录于 2026-10-26
Joshua Wasserman, Herbert Pang, Jiawen Zhu
摘要
Randomized trials have long been the standard for evaluating experimental treatments in medical and pharmacological studies, but the associated time and cost have spurred research into designs that augment randomized studies with data from external sources. Hybrid control arms offer the promise of improved precision-at the possible expense of bias-by forming a control arm from patients randomized to control along with historical controls or patients found in a registry. Although recent reviews have published guidance on minimizing and addressing missing data in clinical trials, the issue has received little attention in the context of trials with a hybrid control arm. We propose novel treatment effect estimators that handle missing outcomes in both the trial and external control datasets, even when the missingness patterns differ between them. These estimators are built from augmented weighted estimating equations and are doubly robust when outcomes are missing at random (MAR). In addition, we introduce a sensitivity analysis using a pattern-mixture model to assess robustness of the effect estimates under violations of the MAR assumption. We compare these estimators to a popular parametric modeling technique through simulation and an application to a real-world trial.
SMMR Longitudinal Data Survival Analysis · 发表于 2026-10-21,收录于 2026-10-26
Avinash Kumar, M S Panwar, Sushil Kumar
摘要
A joint modeling technique for longitudinal data and time-to-event data is advanced statistical methodology to analyze repeated measurements and time-to-event outcomes. In the present article, a shared random effects approach is considered to account for both longitudinal data and survival outcomes. The longitudinal data is submodeled by a linear mixed-effects model that captures the individual response over time. Whereas the time-to-event sub-model uses the generalized exponential distribution as the baseline distribution for the proportional hazard model to incorporate survival times as well as time-independent covariates. To estimate model parameters, the Expectation-Maximization approach has been used, which is effective with latent random effects since random effects are unobservable in nature. In the E-step, the expected log-likelihood is derived based on current parameter estimates, and in the M-step, we maximize this likelihood to update the parameters. To show the efficacy of the proposed approach, an extensive simulation study has been conducted. Furthermore, the model is tested on a real-world medical dataset, heart valve data, to demonstrate its efficacy in analyzing the link between longitudinal responses and survival outcomes.
SiM Clinical Trials · 发表于 2026-10-21,收录于 2026-10-26
Jennifer A Thompson, Brennan C Kahan, Andrew Copas
摘要
BACKGROUND: Methods for estimating marginal effects in cluster randomized trials are limited. We describe and evaluate a cluster-level generalized linear model (CL-GLM) that applies a GLM to cluster-level summaries (means, proportions, or rates) and can estimate differences, ratios, or odds ratios. We focus on odds ratios. METHODS: We conducted a simulation study with 10-50 clusters, intracluster correlation coefficients of 0.05 or 0.15, and a mean cluster size of 120, either fixed or variable. Informative cluster size, where the intervention effect varies with cluster size, was also considered. Control arm prevalence was set to 50%. We compared cluster-level GLMs with normal and gamma distributions and logit link to generalized estimating equations (GEE) and independence estimating equations (IEEs). Weighted and unweighted analyses were examined to target cluster-average and individual-average estimands. RESULTS: CL-GLM with a normal distribution produced unbiased marginal cluster-average odds ratios with appropriate confidence interval coverage. CL-GLM with a gamma distribution resulted in over-coverage. Unweighted IEE produced unbiased marginal individual-average odds ratios with correct coverage. GEE was unbiased when cluster size was uninformative but biased under informative cluster size. Weighted CL-GLM led to under-coverage, whereas weighted IEE resulted in over-coverage. The CL-GLM had similar power to GEE and unweighted IEE. CONCLUSIONS: The cluster-level GLM with a normal distribution provides a robust alternative to estimating equation approaches for marginal odds ratios in cluster randomized trials, with good performance under both informative and uninformative cluster size and no loss of power. We recommend the cluster-level GLM with a normal distribution when the marginal cluster-average estimand is of interest, and unweighted IEEs when the marginal individual-average estimand is of interest.