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 4 页
共 1756 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2026-10-09 到 2026-10-01。
按期刊、按主题看 →
SiM Survival Analysis Nonparametric & Semiparametric Missing Data & Measurement Error · 发表于 2026-10-09,收录于 2026-10-20
Fuyu Guo, Zihan Qian, Wenyi Wang, Fenglei Wang, Bernard Rosner, Molin Wang
摘要
External exposures take time to accumulate and affect human health, leading to a time lag between initial exposure and the manifestation of clinical signs. Susceptible time window analysis aims to describe the relationship between the time since exposure and its effects. To obtain unbiased estimates in susceptible time window analysis, researchers may need to adjust for exposures measured at different times simultaneously. Conventional Cox proportional hazards models often fail in this regard due to the strong multicollinearity between lagged exposure terms. In addition to estimation challenges, missing data in historical exposures present another obstacle for long-term analysis. To address these challenges, we first impose flexible parametric models on the effects across historical exposures by assuming that they follow restricted cubic splines. Based on our proposed restricted cubic spline model, we develop two model selection algorithms to help researchers determine the optimal knot locations in the splines. Additionally, we propose using multiple imputations to handle the missing exposure data. We conducted simulation studies to evaluate the performance of the restricted cubic spline models and the imputation algorithms in various realistic settings. The results show that our model demonstrates strong performance, offering great flexibility and robustness to misspecified maximum length of effect. Compared to the complete-person-time analysis, our multiple imputation algorithm exhibits significantly lower variance, with only a slight increase in bias. Finally, as an illustrative example, we apply our proposed models to analyze the effect of folate consumption on colorectal cancer incidence in the Nurses' Health Study from 1976 to 2016.
SiM Longitudinal Data · 发表于 2026-10-09,收录于 2026-10-20
Keyliane Travassos, Larissa A Matos, Fernanda L Schumacher
摘要
Linear mixed models are often used to analyze studies with repeated measurements over time, where the data present a correlation between observations of the same individual. However, one of the main challenges in using these models is correctly defining the covariance structure associated with error and random effects. To overcome this challenge, we obtain the sandwich variance estimator as a robust alternative to estimate the standard error in the scale mixture of skew-normal linear mixed models (SMSN-LMMs). By definition, SMSN-LMMs can accommodate skewness and heavy tails, and by considering our proposed variance estimator, we increase the model's flexibility by allowing for misspecification of the covariance components. We present the definition of the estimator, its derivation for this model, and its practical implementation. In addition, through simulation studies, we evaluate its performance in different scenarios, demonstrating its ability to provide reliable inferences even when the model or the covariance structure is misspecified.
SiM Testing & Uncertainty · 发表于 2026-10-09,收录于 2026-10-20
Ivair R Silva, Judith C Maro
摘要
Sequential hypothesis testing has been used in randomized clinical trials and in monitoring adverse events after vaccination. Typically, the presumption is that all data that have been previously used in the sequential analysis remain the same, and only new data are added. We call this the "anchoring" assumption. In prelicensing randomized trials, enrollment goals are set based on these sequential hypothesis tests so that a trial may stop if a drug proves either too hazardous or, conversely, so beneficial that it would be unethical to deny the treatment to others. However, in monitoring adverse events after vaccination, data are mostly "secondary use," or real-world data. That is, these data are not gathered for research, but are clinical data that have been repurposed for observational vaccine safety surveillance. As such, the counts of exposed persons and adverse events may be subject to revisions over time. These data changes directly challenge the anchoring assumption upon which most sequential hypothesis testing depends. If these data changes are discovered prior to subsequent hypothesis tests, it is unclear how to account for these changes in subsequent tests. We introduce a solution: "robust alpha spending" to protect overall type I error probability, statistical power, and surveillance time from low-level data instability present in secondary use data sources used for adverse event monitoring.
JASA Latent Variables & Mixtures · 发表于 2026-10-09,收录于 2026-07-31
Javier González-Delgado, Mathis Deronzier, Juan Cortés, Pierre Neuvial
摘要
Recent work by Gao et al. (2024) has laid the foundations for post-clustering inference, establishing a theoretical framework allowing to test for differences between means of estimated clusters. Additionally, they studied the estimation of unknown parameters while controlling the selective type I error. However, their theory was developed for independent observations identically distributed as p-dimensional Gaussian variables, where the parameter estimation could only be performed for spherical covariance matrices. Here, we aim at extending this framework to a more convenient scenario for practical applications, where arbitrary dependence structures between observations and features are allowed. We establish sufficient conditions for extending the setting presented by Gao et al. (2024) to the general dependence framework. Moreover, we assess theoretical conditions allowing the compatible estimation of a covariance matrix. The theory is developed for hierarchical agglomerative clustering algorithms with several types of linkages, and for the k-means algorithm. We illustrate our method with synthetic data and real data of protein structures. *(abstract via OpenAlex)*
JASA High-Dimensional · 发表于 2026-10-09,收录于 2026-10-20
Yilin Zhang, Songshan Yang, Yunan Wu, Lan Wang
摘要
This paper introduces a novel robust high-dimensional inference framework based on partial Gini covariance. We specifically address the practically important problem of quantifying how changes in features affect the conditional mean in high-dimensional linear regression when the random noises are heavy-tailed and possibly asymmetric. Conventional high-dimensional regression inference methods suffer from inaccurate type I errors and reduced power in heavy-tailed contexts. While adaptive Huber regression provides a principled remedy, its effectiveness relies on the careful calibration of robustification parameters that must adapt to sample size, dimensionality, and tail behavior. Our proposed approach leverages the partial Gini covariance to construct a robust statistical inference framework that requires minimal tuning and avoids restrictive moment conditions on error distributions. By circumventing the need for error density estimation, this method enhances both computational feasibility and robustness. Extensive simulations and a real-world data example demonstrate the proposed method’s superior power and robustness over standard high-dimensional inference approaches, such as those based on the debiased Lasso. The asymptotic relative efficiency analysis provides additional theoretical insight on the improved efficiency of the new approach in the heavy-tailed setting. Additionally, the partial Gini covariance extends to the multivariate setting, enabling chi-square testing for a group of coefficients. *(abstract via OpenAlex)*
SiM Causal Inference Bayesian Methods Genetics & Omics · 发表于 2026-10-09,收录于 2026-10-20
Andrew J Grant, Ashish Patel, Stephen Burgess
摘要
Weak instruments are a well‐known limitation for valid causal inference in Mendelian randomization studies. In the single exposure setting, weak instrument bias can be mitigated by selecting genetic instruments which are strongly associated with the exposure according to p ‐value and/or F ‐statistic thresholds. However, in the multi‐exposure setting, genetic instruments may be strongly associated with an exposure but weakly associated with it conditional on all other exposures in the analysis. It is therefore more difficult to guarantee conditionally strong instruments in multivariable Mendelian randomization. Weak instrument bias can be mitigated using modeling approaches; however, there are fewer methods for doing this in the multivariable case compared with the single exposure case. In this paper, we consider a method for mitigating weak instrument bias in multivariable Mendelian randomization using a Bayesian framework: MVMR‐Pony. We compare this method with existing frequentist methods. We show using simulation studies that the MVMR‐Pony method outperforms the frequentist approaches with respect to bias, coverage, Type I error rates, and power, across settings where weak instrument bias arises due to correlated genetic effects, measurement error, and mediation.
SMMR Survival Analysis Machine Learning Networks & Graphs · 发表于 2026-10-09
Junyao Ren, Shishun Zhao, Dianliang Deng, Shuwei Li
摘要
Mixture cure rate models represent a valuable tool to delineate the presence of a cured subgroup in the entire population under study. Among the vast literature, a popular research line is to regress the probability of an individual being cured on covariates through a parametric model, such as the logistic model. However, when the assumed parametric model is mis-specified, one could only attain a biased parameter estimate. In this work, motivated by the robustness and powerful approximation ability of neural networks, we offer a flexible mixture cure rate modeling approach for analyzing interval-censored data, which arise frequently in many scientific fields involving periodic follow-up or cross-sectional screening. In particular, we utilize the artificial neural network to model the cured probability and the proportional hazards model to characterize the latent event time distribution related to uncured individuals. After approximating the cumulative baseline hazard function with monotone splines, we develop a stable expectation-maximization algorithm coupled with a multiple imputation strategy and resilient backpropagation to locate the sieve maximum likelihood estimator. We establish the consistency and convergence rate of the proposed sieve estimator, as well as the asymptotic normality and semiparametric efficiency of the regression parameter estimator. Simulation experiments demonstrate that our proposed method works well and substantially outperforms the comparative methods. We then apply the proposed method to a real-world data set, revealing new findings and a better predictive performance.
JASA Meta-Analysis & Data Integration · 发表于 2026-10-09,收录于 2026-07-24
Not available
摘要
Multi-center research has strengthened the generalizability and reproducibility of scientific findings, yet multi-site inference in modern distributed research networks remains constrained by privacy regulations, heterogeneous site characteristics, and the logistical burden of human-in-the-loop, multi-round communication. In this work, we distill these operational constraints into four mathematical requirements for practical distributed inference and introduce MOSAiC, a unified meta-algorithmic framework that addresses them simultaneously. MOSAiC leverages tensor-train tools from scientific computing to reframe distributed learning as a problem of multivariate function approximation, enabling one-shot communication, robustness to small or irregular sites, accuracy comparable to pooled analysis, and analytic submodel flexibility without additional queries. We demonstrate the feasibility and scalability of the approach through deployments across four institutions and applications in drug relabeling, drug repurposing, and post-market safety surveillance. Beyond these results, MOSAiC offers an operationally grounded foundation for designing next-generation federated learning methods — one that aligns statistical rigor with the operational rhythms of real-world scientific collaboration and opens new directions for methodological innovation. *(abstract via OpenAlex)*
SiM Longitudinal Data Survival Analysis Spatial & Time Series · 发表于 2026-10-09,收录于 2026-10-20
Yue Zhan, Cheng Zheng, Ying Zhang
摘要
Huntington's disease (HD) is an autosomal dominant neurodegenerative disorder characterized by motor dysfunction, psychiatric disturbances, and cognitive decline. The onset of HD is marked by severe motor impairment, which may be predicted by prior cognitive decline and, in turn, exacerbate cognitive deficits. Clinical data, however, are often collected at discrete time points, so the timing of disease onset is subject to interval censoring. To address the challenges posed by such data, we develop a joint model for multivariate longitudinal biomarkers with a change point anchored at an interval-censored event time. The model simultaneously assesses the effects of longitudinal biomarkers on the event time and the changes in biomarker trajectories following the event. We conduct a comprehensive simulation study to demonstrate the finite-sample performance of the proposed method for statistical inference based on the asymptotic normality theorem. Finally, we apply the method to PREDICT-HD, a multisite observational cohort study of prodromal HD individuals, to ascertain how cognitive impairment and motor dysfunction interact during disease progression.
JASA Machine Learning Networks & Graphs · 发表于 2026-10-09,收录于 2026-10-20
Jiping Wang, Tong Wang, Shuangge Ma
摘要
In outcomes research, clinical treatment measures are routinely analyzed. Most existing studies focus on individual diseases, and increasing recent evidence suggests that clinical treatment measures of different diseases can be “correlated” (or interconnected). Existing studies on such interconnections are usually limited to a small number of diseases. In this study, for U.S. older patients with common cancers, we complement the existing literature and conduct a pan-disease analysis of interconnections in clinical treatment measures. This effort can assist in better managing and planning healthcare resources. We mine SEER-Medicare for subjects diagnosed between January 2004 and December 2017 with esophageal, colorectal, liver, pancreatic, oral cavity/pharyngeal, lung, breast, prostate, renal/kidney, and bladder cancers. In the inpatient/outpatient treatment settings, a total of 113/168 diseases are analyzed. To accommodate unique data characteristics and our analysis goals, we develop a deep neural network (DNN)-based estimation approach, which adopts an additive two-part loss function to accommodate zero inflation, uses DNNs to capture nonlinearity, and applies penalization to identify interconnections. Extensive statistical and numerical investigations are conducted. In the data analysis, sensible findings are made for individual cancers, and across-cancer and across-setting analyses are also conducted. 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 2026-10-09,收录于 2026-10-20
Chen Cheng, Hilal Asi, John Duchi
摘要
The construction of most supervised learning datasets revolves around collecting multiple labels for each instance, then aggregating the labels to form a type of “true” label. We question the wisdom of this pipeline by developing a (stylized) theoretical model of this process and analyzing its statistical consequences, showing how access to non-aggregated label information can make training well-calibrated models more feasible than it is with cleaned labels. The entire story, however, is subtle, and the contrasts between aggregated and fuller label information depend on the particulars of the problem, where estimators that use aggregated information exhibit robust but slower rates of convergence, while estimators that can effectively leverage all labels converge more quickly if they have fidelity to (or can learn) the true labeling process. The theory makes several predictions for real-world datasets, including when non-aggregate labels should improve learning performance, which we test to corroborate the validity of our predictions. *(abstract via OpenAlex)*
SiM Survival Analysis Clinical Trials · 发表于 2026-10-09,收录于 2026-10-20
Susan Halabi, Lu Liu, Chenxi Yu, Yuan Wu
摘要
We propose a novel method that simultaneously determines the sample size for testing two survival probabilities at a pre-specified time while guaranteeing Type I error control in both fixed and group-sequential trial designs. Simulations across varying hypothesized differences, failure distributions, censoring proportions, and nominal powers demonstrate consistent performance, while interim analyses highlight maintained nominal Type I error control and increased power at each look relative to the standard binomial approach, regardless of the underlying failure time distribution or spending function. Importantly, our method is especially useful for evaluating survival outcomes at a fixed time in randomized trials where one treatment arm includes neoadjuvant therapy prior to surgery while the other involves surgery alone. Furthermore, it is advantageous when the proportional hazards assumption is not satisfied, as often occurs in immunotherapy trials with delayed or time-varying treatment effects or crossing survival curves. The method is also applicable to randomized Phase II trials, where smaller sample sizes and the use of intermediate or surrogate time-to-event endpoints demand efficient data use and robust error control. We illustrate the approach with motivating examples in renal and prostate cancer. An accompanying R Shiny application enables investigators to compute sample sizes interactively, facilitating practical trial planning in diverse settings.
Biometrics Survival Analysis · 发表于 2026-10-09,收录于 2026-10-20
Arron Gosnell, Evangelos Evangelou
摘要
With the proliferation of screening tools for chemical testing, it is now possible to create vast databases of chemicals easily. However, rigorous statistical methodologies employed to analyze these databases are in their infancy, and further development to facilitate chemical discovery is imperative. In this paper, we address the challenge of predicting an organic solvent's water pollution class based on its chemical structure. We hypothesize that solvents with similar chemical structure are more likely to belong in the same hazard class. This chemical similarity is measured by the Tanimoto distance, a non-Euclidean metric on the chemical space. To incorporate the similarity between chemical structures in the model, we propose a Gaussian process model on the chemical space, with the kernel being a function of the Tanimoto distance. A novel feature of the proposed model is the inclusion of a scale parameter in the kernel, which controls the strength of the correlation between compounds and offers additional flexibility. We find that accounting for correlation between chemical compounds substantially improves predictive performance over the uncorrelated model, where compound similarity is unaccounted for. Our model also compares favorably against other established models in the literature. Furthermore, we present a genetic algorithm designed to identify important features to a compound's efficacy, with the goal of facilitating chemical discovery. The algorithm operates based on two criteria derived from the proposed model. Simulation studies are conducted to demonstrate the suitability of the proposed methods.
SiM Testing & Uncertainty Missing Data & Measurement Error Diagnostics & Prediction · 发表于 2026-10-09,收录于 2026-10-20
Aryana Arsham, Paul S Albert, Mark P Little
摘要
Bias-adjusted estimators are proposed for logistic regression parameters under a mixture measurement error model. In observational studies, collecting the true exposure for all units is often infeasible. This has led to the use of prediction models to estimate the true exposure. However, these estimates are error-prone, introducing measurement error. Measurement error in statistical analysis can result in biased parameter estimates and unreliable conclusions. The method developed in this work uses a mixture error model that accounts for both an error-prone predicted exposure and true exposure. The proposed bias-adjusted approach uses bootstrap estimators of the mixture error model parameters, a user-specified prediction model, and a linear regression calibration function. The approach applies to cohort designs under various exposure distributions when measurement error model parameters are unknown. A Wald confidence interval is constructed using the developed bias-adjusted estimators and their bootstrapped standard errors. The proposed method is evaluated in a simulation study and an observational study in which a logistic regression model relates hypertension incidence to glycohemoglobin exposure. Numerical results demonstrate accurate point estimates and coverage probabilities close to the nominal level. We further investigate the effects of selecting an incorrect prediction model on the proposed approach. The results indicate that using an oversimplified predictive model can lead to severe biases. Additionally, we show that even flexible machine learning models may offer lower accuracy and precision than correctly specified parametric models. Diagnostics for assessing model assumptions are provided in the Supporting Information and illustrated for the motivating example.
SMMR Survival Analysis Bayesian Methods · 发表于 2026-10-09
Matilda Pitt, Robert JB Goudie
摘要
Retrospective cohorts can be extracted from electronic health records (EHRs) to study prevalence, time until disease or event occurrence, and cure proportion in real-world scenarios. However, EHRs are collected for patient care rather than research, and thus typically have complexities, such as patients with missing baseline disease status. Prevalence incidence (PI) models, which use a two-component mixture model to account for this missing data, have been proposed. However, PI models are biased in settings in which some individuals will never experience the endpoint (they are ‘cured’). To address this, we propose a prevalence-incidence-cure (PIC) model, a three-component mixture model that combines the PI model framework with a cure model. Our PIC model enables estimation of the prevalence rate, time-to-incidence, and the cure proportion, and allows for covariate effects on each component. We adopt a Bayesian inference approach, and focus on the interpretability of the prior. We show in a simulation study that the PIC model has smaller bias than PI and cure models for the survival probability; and we compare inference under vague, informative and misspecified priors. We illustrate our PIC model using a dataset of 1964 patients who underwent treatment for diabetic macular oedema, and demonstrate improved fit under our model.
JASA Causal Inference · 发表于 2026-10-08,收录于 2026-10-20
Samuel D. Pimentel, Ruoqi Yu
摘要
Matching is an appealing way to design observational studies because it mimics the data structure produced by stratified randomized trials, pairing treated individuals with similar controls. After matching, inference is often conducted using methods tailored for stratified randomized trials in which treatments are permuted within matched pairs. However, in observational studies, matched pairs are not predetermined before treatment; instead, they are constructed based on observed treatment status. This introduces a challenge as the permutation distributions used in standard inference methods do not account for the possibility that permuting treatments might lead to a different selection of matched pairs (Z-dependence). To address this issue, we propose a novel and computationally efficient algorithm that characterizes and enables sampling from the correct conditional distribution of treatment after an optimal propensity score matching, accounting for Z-dependence. We show how this new procedure, called match-adaptive randomization inference, corrects for an anticonservative result in a well-known observational study investigating the impact of hormone replacement theory (HRT) on coronary heart disease and corroborates experimental findings about heterogeneous effects of HRT across different ages of initiation in women. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*
Biometrics Meta-Analysis & Data Integration · 发表于 2026-10-08,收录于 2026-10-20
Hisashi Noma
摘要
In meta-analyses of binary outcomes, two-stage methods based on the DerSimonian-Laird-type random-effects model have been widely used, but these methods fix the within-study variances to their asymptotic estimates and rely on normality assumptions of the outcome measures. Such strict and often unrealistic assumptions can yield biased or inefficient inference. Likelihood-based approaches such as the binomial-normal generalized linear mixed model and the hypergeometric-normal (HGN) model alleviate some of these issues but are intrinsically restricted to the odds-ratio scale, limiting interpretability and precluding direct estimation of the risk-ratio-a more clinically meaningful measure. To overcome these limitations, we propose a unified likelihood-based framework that enables direct estimation of the risk-ratio under the HGN formulation. The proposed method introduces a pseudo-data augmentation approach that preserves the conditional likelihood structure of the HGN model while yielding an unbiased estimating function for the log-risk-ratio. Higher-order inference is achieved via jackknife bias and variance corrections, ensuring accurate coverage even in small or sparse datasets. Simulation studies confirm nearly unbiased estimation and well-calibrated confidence intervals across a wide range of conditions. Applications to published meta-analyses demonstrate that this approach yields interpretable, coherent, and computationally tractable inference.
Biometrics Meta-Analysis & Data Integration Survival Analysis Bayesian Methods · 发表于 2026-10-08,收录于 2026-10-20
Daisuke Yoneoka, Takayuki Kawashima, Yuta Tanoue, Shuhei Nomura, Akifumi Eguchi
摘要
Accurate estimation of incubation period is fundamental to surveillance for infectious diseases. This task faces two major challenges: (1) surveillance data are often coarsely observed, comprising interval- and right-censored observations, and (2) patient-level data are frequently fragmented across multiple facilities and cannot be pooled due to privacy constraints. While traditional meta-analysis (MA) aggregates summary statistics, its one-shot nature limits the ability to fit complex hierarchical models. We propose a novel federated Bayesian framework, Federated Stochastic Variational Inference (Fed-SVI), which extends the summary-statistic philosophy of MA into an iterative, model-based federated approach. Our method fits a hierarchical lognormal model, explicitly modeling heterogeneity while estimating a global distribution. To fit this model privately, Fed-SVI iteratively exchanges summary statistics (local variational parameters). Each facility analytically computes these local parameters, rigorously incorporating the likelihood contributions of coarsely-observed data via expectations of the truncated normal distribution. A central server aggregates these parameters using stochastic natural-gradient ascent, enabling efficient information borrowing and mini-batch updates. Through extensive simulations, we demonstrate that Fed-SVI achieves accuracy nearly identical to a fully centralized analysis, while substantially outperforming naive federated or MA methods that discard censored information. We apply Fed-SVI to a distributed COVID-19 line list from 177 facilities, yielding results consistent with centralized estimates. Fed-SVI provides a scalable, accurate, and privacy-preserving solution for robust surveillance in modern data networks.
Biometrics Causal Inference Nonparametric & Semiparametric Machine Learning · 发表于 2026-10-08,收录于 2026-10-20
Fan Xia, Sang Kyu Lee, Seonjin Kim, Kelly Blum, Hyokyoung G Hong, Alison J Canchola, Mi-Ok Kim
摘要
This study proposes a nonparametric decomposition framework to quantify and prioritize the impact of multiple intervenable mediators contributing to disparities in health outcomes. Identifying modifiable components of health disparities is essential for informing policy and guiding resource allocation, but several methodological challenges remain. These include formulating hypothetical interventions for immutable characteristics such as race or ethnicity, mitigating bias from potentially incompatible statistical models used to represent causal pathways, and evaluating multiple mediators for their joint and individual effects. To address these challenges, we integrate recent advances in health disparities research with modern machine learning techniques. A key assumption, common to causal disparity frameworks, is the absence of unmeasured confounding in the mediator-outcome relationship. We extend existing methods in mediation analysis and introduce a nonparametric sensitivity analysis that provides sharp bounds on modifiable disparity estimates in the presence of unmeasured confounding. Together, these methods offer a robust and flexible approach to causal disparity analysis, enabling the identification of high-impact, actionable interventions. We demonstrate the utility of the methods in the context of colorectal cancer survival disparities, a setting where persistent differences across racial, ethnic, and underserved populations in the United States underscore the urgent need for analytic strategies that uncover effective intervention points.
Econometrica Econometrics & Economics · 发表于 2026-10-08
Stéphane Bonhomme, Koen Jochmans, Martin Weidner
摘要
*(arXiv preprint abstract, arXiv:2412.10304, verbatim; the journal version may differ)* A popular approach to perform inference on a target parameter in the presence of nuisance parameters is to construct estimating equations that are orthogonal to the nuisance parameters, in the sense that their expected first derivative is zero. Such first-order orthogonalization allows the estimator of the nuisance parameters to converge at a slower-than-parametric rate. It may, however, not suffice when the nuisance parameters are very imprecisely estimated. Leading examples are models for panel and network data that feature fixed effects. In this paper, we show how, in the conditional-likelihood setting, estimating equations can be constructed that are orthogonal to any chosen order q, in that their leading q expected derivatives are zero. This yields estimators of target parameters that are unaffected by the presence of nuisance parameters to order q. In an empirical illustration, we apply our method to a fixed-effect model of team production.
JASA Meta-Analysis & Data Integration · 发表于 2026-10-07,收录于 2026-10-20
Lauren J. Beesley, Dave Osthus, Kelly R. Moran, Madeline A. Stricklin, Grant David Meadors, Thomas K. Kim, Sung Jun Noh, Nehpreet K. Walia, Paul H. Janzen, Eric J. Zirnstein, Brian P. Weaver, Daniel B. Reisenfeld
摘要
NASA’s Interstellar Boundary Explorer (IBEX) satellite collects data on energetic neutral atoms (ENAs) that can provide insight into the heliosphere boundary between our solar system and interstellar space. Using these data, scientists can construct maps of the ENA intensities (often, expressed in terms of flux) observed in all directions. The ENA flux observed in these maps is believed to come from at least two distinct sources: one source which manifests as a ribbon of concentrated ENA flux and one source (or possibly several) that results in a smoothly-varying globally-distributed flux. Each ENA source type and its corresponding ENA intensity map is of separate scientific interest.In this paper, we develop statistical algorithms for separating the total ENA intensity maps into two source-specific maps (ribbon and globally-distributed flux) and estimating corresponding uncertainty. Key advantages of the proposed method include enhanced model flexibility and improved propagation of estimation uncertainty. We evaluate the proposed methods on simulated data designed to mimic realistic data settings. We also propose new methods for estimating the center of the near-elliptical ribbon in the sky, which can be used in the future to study the location and variation of the local interstellar magnetic field. *(abstract via OpenAlex)*
JASA Latent Variables & Mixtures Optimization & Learning Theory · 发表于 2026-10-07,收录于 2026-10-08
Kwong Yu Chong, Long Feng
摘要
*(arXiv preprint abstract, arXiv:2603.04223, verbatim; the journal version may differ)* We introduce Latent Space Distribution Matching (LSDM), a novel framework for semi-supervised generative modeling of conditional distributions. LSDM operates in two stages: (i) learning a low-dimensional latent space from both paired and unpaired data, and (ii) performing joint distribution matching in this space via the 1-Wasserstein distance, using only paired data. This two-step approach minimizes an upper bound on the 1-Wasserstein distance between joint distributions, reducing reliance on scarce paired samples while enabling fast one-step generation. Theoretically, we establish non-asymptotic error bounds and demonstrate a key benefit of unpaired data: enhanced geometric fidelity in generated outputs. Furthermore, by extending the scope of its two core steps, LSDM provides a coherent statistical perspective that connects to a broad class of latent-space approaches. Notably, Latent Diffusion Models (LDMs) can be viewed as a variant of LSDM, in which joint distribution matching is achieved indirectly via score matching. Consequently, our results also provide theoretical insights into the consistency of LDMs. Empirical evaluations on real-world image tasks, including class-conditional generation and image super-resolution, demonstrate the effectiveness of LSDM in leveraging unpaired data to enhance generation quality.
Biometrics Meta-Analysis & Data Integration Testing & Uncertainty · 发表于 2026-10-07,收录于 2026-10-20
David Kronthaler, Leonhard Held
摘要
Statistical inference about the average effect in random-effects meta-analysis has been considered insufficient in the presence of substantial between-study heterogeneity. Predictive distributions are well suited for quantifying heterogeneity since they are interpretable on the effect scale and provide clinically relevant information about future events. We construct predictive distributions accounting for uncertainty through confidence distributions from Edgington's p-value combination method and the generalized heterogeneity statistic. Simulation results suggest that 95% prediction intervals typically achieve nominal coverage when more than three studies are available and effectively reflect skewness of effect estimates in scenarios with 20 or fewer studies. Formulations that ignore uncertainty in heterogeneity estimation typically fail to achieve correct coverage, underscoring the need for this adjustment in random-effects meta-analysis.
JASA Causal Inference High-Dimensional · 发表于 2026-10-07,收录于 2026-10-08
Xiufan Yu, Ken Kelley
摘要
Mediation models are widely used in many scientific areas to understand how an exposure variable impacts an outcome variable through an intermediate variable. In this article, we develop powerful inferential methods to analyze mediation effects in high-dimensional contexts with intercorrelated potential mediators in models with linear and generalized linear outcomes. We draw attention to heterogeneous mediation, and show that existing testing approaches based on the total indirect effect are often underpowered when the mediation effects are not homogeneous. To overcome this limitation, we investigate the root cause and propose new solutions. The proposed tests are equipped with expanded capability to handle heterogeneous and even contrasting mediation effects, and have substantial increases in statistical power compared to existing methods, importantly, without compromising the Type I error rate. We demonstrate robust test performance under different mediation patterns by providing rigorous theoretical justifications and comprehensive numerical studies. The practicality is further demonstrated using a real-data application investigating the mediating role of healthcare expenditures in the relationship between economic growth and public health outcomes. An R package POEM is developed to facilitate practical implementation. *(abstract via OpenAlex)*
Biometrics Survival Analysis Nonparametric & Semiparametric · 发表于 2026-10-07,收录于 2026-10-20
Xingyu Zhou, Wen Su, Changyu Liu, Yuling Jiao, Xingqiu Zhao, Jian Huang
摘要
We propose a deep generative approach to nonparametric estimation of conditional survival and hazard functions with censored data. The key idea of the proposed method is to first learn a conditional generator for the joint conditional distribution of the observed time and censoring indicator given covariates, and then construct the Kaplan-Meier and Nelson-Aalen estimators based on this conditional generator for conditional hazard and survival functions. Our method combines ideas from the recently developed deep generative learning and classical nonparametric estimation in survival analysis. We establish the convergence properties of the generative nonparametric estimators. Our numerical studies validate the proposed method.
Biometrics Machine Learning Bayesian Methods Networks & Graphs · 发表于 2026-10-07,收录于 2026-10-20
Xinyuan Tian, Yuanshuang Guo, Shengxian Ding, Yifei Zhang, Denise Esserman, Yize Zhao
摘要
Recent neuroimaging studies have revealed that brain functional networks exhibit temporal fluctuations during cognitive tasks, sparking growing interest in dynamic functional connectivity (FC). However, few existing methods effectively incorporate temporal dynamics when examining the relationship between brain connectivity and cognition. In this work, we propose a Bayesian deep learning framework that decomposes FC into dynamic and stationary components, thereby disentangling transient fluctuations from stable network structures. By leveraging deep kernel learning, the model flexibly represents transient functions as continuous processes, allowing the data to guide the discovery of meaningful temporal patterns beyond conventional assumptions. Building on this representation, the dynamic component is modeled via a thresholded Gaussian process, which captures smooth temporal variations while inducing sparsity to highlight significant temporal signals. Simultaneously, the stationary component is modeled using a deep neural network with an integrated thresholding layer, enabling the learning of complex nonlinear relationships and enhancing interpretability by emphasizing key connections. This integrative approach balances model flexibility, temporal resolution, and interpretability, providing a powerful tool for predicting cognitive outcomes from functional neuroimaging data. We demonstrate that our method outperforms existing approaches in both prediction accuracy and feature selection through extensive simulations. Applied to data from the Adolescent Brain Cognitive Development study, our framework uncovers dynamic brain signals during cognitive tasks that significantly contribute to variations in cognitive performance.
JASA Longitudinal Data Clinical Trials · 发表于 2026-10-06,收录于 2026-10-08
Yang Liu, Lucy Xia, Feifang Hu
摘要
Recent clinical studies increasingly involve a long factor with many levels, e.g., investigation sites, resulting in a large number of strata that must be accounted for either through design or subsequent analysis. This complication has prompted concerns from the U.S. Food and Drug Administration (FDA, 2023) regarding the adequacy of standard statistical methods, whose performance may deteriorate, and their properties become unclear when the number of strata is relatively large. In this work, we offer a first-time rigorous solution by employing mixed-effect models in covariate-adaptive randomized experiments. We show that the mixed-effect estimate achieves lower variance in treatment effect estimation than its fixed-effect counterpart in the presence of the long factor with many levels. This variance reduction is attributable to marginal imbalances induced by the randomization procedure, suggesting that designs promoting finer covariate balance lead to more efficient inference and increased statistical power. Furthermore, we demonstrate that, as the sample size grows, the mixed- and fixed-effect estimators become asymptotically equivalent. Our theoretical findings are validated through simulations and a clinical trial case study, and provide new insights on the design and analysis of clinical trials in the presence of a large number of strata. *(abstract via OpenAlex)*
JASA Networks & Graphs Meta-Analysis & Data Integration · 发表于 2026-10-06,收录于 2026-10-20
Yongqin Qiu, Xinyu Zhang
摘要
Link prediction in multilayer networks is a key challenge in applications such as recommendation systems and protein-protein interaction prediction. While many techniques have been developed, most rely on assumptions about shared structures and require access to raw auxiliary data, limiting their practicality. To address these issues, we propose a novel transfer learning framework for multilayer networks using a bi-level model averaging method. A K-fold cross-validation criterion based on edges is used to automatically weight inter-layer and intra-layer candidate models. This enables the transfer of information from auxiliary layers while mitigating model uncertainty, even without prior knowledge of shared structures. Theoretically, we prove the optimality and weight convergence of our method under mild conditions. Computationally, our framework supports parallel processing across multiple servers without requiring raw data transmission, making it applicable under data-sharing constraints. Simulations show our method outperforms others in predictive accuracy and robustness. We further demonstrate its practical value through two real-world recommendation system applications. *(abstract via OpenAlex)*
JASA Networks & Graphs Spatial & Time Series Bayesian Methods · 发表于 2026-10-05
J. Brandon Carter, Catherine A. Calder
摘要
Current approaches for modeling discrete-valued outcomes associated with spatially-dependent areal units incur computational and theoretical challenges, especially in the Bayesian setting when full posterior inference is desired. As an alternative, we propose a novel statistical modeling framework for this data setting, namely a mixture of directed graphical models (MDGMs). The components of the mixture, directed graphical models, can be represented by directed acyclic graphs (DAGs) and are computationally quick to evaluate. The DAGs representing the mixture components are selected to correspond to an undirected graphical representation of an assumed spatial contiguity/dependence structure of the areal units, which underlies the specification of traditional modeling approaches for discrete spatial processes such as Markov random fields (MRFs). Notably, the MDGM is not proposed as an approximation to an MRF, but rather as an alternative that provides valid posterior inference while being computationally faster than exact MRF inference and more principled than the pseudo-likelihood approximation (aMRF) commonly used in practice. We introduce the concept of compatibility to show how an undirected graph can be used as a template for the dependencies between areal units to create sets of DAGs which, as a collection, preserve the dependencies represented in the template undirected graph. Lastly, we compare highlighted classes of MDGMs to MRFs and a popular Bayesian MRF model approximation used in high-dimensional settings in a series of simulations and an analysis of ecometrics data collected as part of the Adolescent Health and Development in Context Study. *(abstract via OpenAlex)*
AoS Networks & Graphs Machine Learning · 发表于 2026-10-01,收录于 2026-10-20
Huiqin Li, Guangming Pan, Yanqing Yin
摘要
We study the spectral properties of the Jacobian matrix in artificial neural networks and establish both its limiting spectral distribution and the second-order fluctuations of its spectrum. Our analysis uncovers fundamental phenomena, including universality properties with respect to the weight distribution and a phase transition in spectral behavior driven by the choice of activation function. Leveraging tools from random matrix theory, we first analyze a class of nonlinear covariance ensembles, which may be of independent interest in high-dimensional statistics, and subsequently characterize the spectral fluctuations of the Jacobian matrix, highlighting their dependence on weight initialization, layer widths, activation functions and network depth. These findings provide finite-width distributional calibration for Jacobian-based stability diagnostics and quantify initialization-induced uncertainty in signal propagation, gradient stability and local sensitivity of high-dimensional neural networks.
AoS Networks & Graphs · 发表于 2026-10-01,收录于 2026-09-03
Jianqing Fan, Yingying Fan, Jinchi Lv, Fan Yang
摘要
*(publisher's abstract as deposited in Crossref, verbatim)* Large-scale network inference with uncertainty quantification has important applications in natural, social and medical sciences. The recent work of Fan et al. (J. R. Stat. Soc. Ser. B. Stat. Methodol. 84 (2022) 630–653) introduced a general framework of statistical inference on membership profiles in large networks (SIMPLE) for testing the sharp null hypothesis that a pair of given nodes share the same membership profile. In real applications, there are often groups of nodes under investigation that may share similar membership profiles in the presence of relatively weaker signals than the setting considered in SIMPLE. To address these practical challenges, in this paper, we propose a SIMPLE method with random coupling (SIMPLE-RC) for testing the nonsharp composite null hypothesis that a group of given nodes shares similar (not necessarily identical) membership profiles under weaker signals. Utilizing the idea of random coupling, we construct our test as the maximum of the SIMPLE tests for subsampled node pairs from the group. Such a technique significantly reduces the correlation among individual SIMPLE tests while largely maintaining the power, enabling delicate analysis of the asymptotic distributions of the SIMPLE-RC test. Our method and theory cover both the cases with and without node degree heterogeneity. These new theoretical developments are empowered by a second-order expansion of spiked eigenvectors under the ℓ∞-norm, built upon our work for random matrices with weak spikes. Our theoretical results and the practical advantages of the newly proposed method are demonstrated through simulations and real data examples.
AoS Causal Inference · 发表于 2026-10-01,收录于 2026-08-20
Peng Ding, Zhichao Jiang
摘要
*(publisher's abstract as deposited in Crossref, verbatim)* Crossover designs randomly assign each unit to receive a sequence of treatments. By comparing outcomes within the same unit, these designs can effectively eliminate between-unit variation and facilitate the identification of both instantaneous effects of current treatments and carryover effects from past treatments. They are widely used in traditional biomedical studies and are increasingly adopted in modern digital platforms. However, standard analyses of crossover designs often rely on strong parametric models, making inference vulnerable to model misspecification. This paper adopts the design-based framework to analyze general crossover designs. We make two main contributions. First, we use potential outcomes to formally define the causal estimands and assumptions on the data-generating process. For any given type of crossover design and assumptions on potential outcomes, we outline a procedure for identification and estimation, emphasizing the central role of the treatment assignment mechanism in the design-based inference. Second, we unify the analysis of crossover designs using least squares, with restrictions on the coefficients and weights on the units. Based on the theory, we recommend the specification of the regression function, weighting scheme, and coefficient restrictions to assess identifiability, construct efficient estimators, and conduct inference in a unified fashion. Crucially, the least squares procedure is simple to implement, and yields not only consistent and efficient point estimates but also valid inference even when the working regression model is misspecified.
AoS Networks & Graphs · 发表于 2026-10-01,收录于 2026-10-20
Rahul Raphael Kanekar
摘要
In this paper, we study the problem of testing the equality of two multivariate distributions. One class of tests used for this purpose utilizes geometric graphs constructed using interpoint distances. So far, the asymptotic theory of these tests applies only to graphs, which fall under the stabilizing graphs framework of Penrose and Yukich (Ann. Appl. Probab. 13 (2003) 277–303). We study the case of the K-nearest neighbors graph where K=kN increases with the sample size, which does not fall under the stabilizing graphs framework. Our main result gives detection thresholds for this test in parametrized families when kN=o(N1/4), thus extending the family of graphs where the theoretical behavior is known. We propose a 2-sided version of the test, which removes an exponent gap that plagues the 1-sided test. Our result also shows that increasing the number of nearest neighbors improves the power of the test. This provides a theoretical justification for using denser graphs in testing the equality of two distributions.
AoS Meta-Analysis & Data Integration · 发表于 2026-10-01,收录于 2026-10-20
Yinrui Sun, Yin Xia
摘要
Distributed learning offers a practical solution for the integrative analysis of multisource datasets, especially under privacy or communication constraints. However, addressing prospective distributional heterogeneity and ensuring communication efficiency pose significant challenges on distributed statistical analysis. In this article we focus on integrative estimation of distributed heterogeneous precision matrices, a crucial task related to joint precision matrix estimation where computation-efficient algorithms and statistical optimality theories are still underdeveloped. To tackle these challenges, we introduce a novel HEterogeneity-adjusted Aggregating and Thresholding (HEAT) approach for distributed integrative estimation. HEAT is designed to be both communication- and computation-efficient, and we demonstrate its statistical optimality by establishing the convergence rates and the corresponding minimax lower bounds under various integrative losses. To enhance the optimality of HEAT, we further propose an iterative HEAT (IteHEAT) approach. By iteratively refining the higher-order errors of HEAT estimators through multi-round communications, IteHEAT achieves geometric contraction rates of convergence. Extensive simulations and real data applications validate the numerical performance of HEAT and IteHEAT methods.
AoS Optimization & Learning Theory Machine Learning · 发表于 2026-10-01,收录于 2026-10-20
Nabarun Deb, Tengyuan Liang
摘要
We introduce a novel generative modeling framework based on a discretized parabolic Monge–Ampère PDE, which emerges as a continuous limit of the Sinkhorn algorithm commonly used in optimal transport. Our method performs iterative refinement in the space of Brenier maps using a mirror gradient descent step. We establish theoretical guarantees for generative modeling through the lens of no-regret analysis, demonstrating that the iterates converge to the optimal Brenier map under a variety of step-size schedules. As a technical contribution, we derive a new Evolution Variational Inequality tailored to the parabolic Monge–Ampère PDE, connecting geometry, transportation cost and regret. Our framework accommodates nonlog-concave target distributions, constructs an optimal sampling process via the Brenier map, and integrates favorable learning techniques from generative adversarial networks and score-based diffusion models. As direct applications, we illustrate how our theory paves new pathways for generative modeling and variational inference.
AoS High-Dimensional · 发表于 2026-10-01,收录于 2026-09-03
Kiana Asgari, Andrea Montanari, Basil Saeed
摘要
*(arXiv preprint abstract, arXiv:2502.01953, verbatim; the journal version may differ)* We consider a general model for high-dimensional empirical risk minimization whereby the data $\mathbf{x}_i$ are $d$-dimensional Gaussian vectors, the model is parametrized by $\mathbf{\Theta}\in\mathbb{R}^{d\times k}$, and the loss depends on the data via the projection $\mathbf{\Theta}^\mathsf{T}\mathbf{x}_i$. This setting covers as special cases classical statistics methods (e.g. multinomial regression and other generalized linear models), but also two-layer fully connected neural networks with $k$ hidden neurons. We use the Kac-Rice formula from Gaussian process theory to derive a bound on the expected number of local minima of this empirical risk, under the proportional asymptotics in which $n,d\to\infty$, with $n\asymp d$. Via Markov's inequality, this bound allows to determine the positions of these minimizers (with exponential deviation bounds) and hence derive sharp asymptotics on the estimation and prediction error. As a special case, we apply our characterization to convex losses. We show that our approach is tight and allows to prove previously conjectured results. In addition, we characterize the spectrum of the Hessian at the minimizer. A companion paper applies our general result to non-convex examples.
AoS Causal Inference Spatial & Time Series · 发表于 2026-10-01,收录于 2026-10-08
Abhirup Datta, Michael L. Stein
摘要
*(publisher's abstract as deposited in Crossref, verbatim)* The problem of estimating the slope parameter in regression between two spatial processes under confounding by an unmeasured spatial process has received widespread attention in the recent statistical literature. Yet, a fundamental question remains unresolved: when is this slope consistently estimable under spatial confounding, with existing insights being largely empirical or estimator-specific. In this manuscript, we characterize conditions for consistent estimability of the regression slope between Gaussian random fields (GRFs), the common stochastic model for spatial processes, under spatial confounding. Under fixed-domain (infill) asymptotics, we give sufficient conditions for consistent estimability in terms of the smoothness or local behavior of the exposure and confounder processes. When estimability holds, we provide consistent estimators of the slope using local differencing (taking discrete differences or Laplacians of the processes of suitable order). Using functional analysis results on Paley–Wiener spaces, we then provide an easy-to-verify necessary condition for consistent estimability of the slope in terms of the relative spectral tail decays of the confounder and exposure. As a by-product, we establish a novel and general spectral condition on the equivalence of measures on the paths of multivariate GRFs with component fields of varying smoothnesses, a result of independent importance. Our estimability results or estimators do not rely on specific parametric models for the covariance functions. We show that for many covariance classes like the Matérn, power-exponential, generalized Cauchy and coregionalization families, the necessary and sufficient conditions become identical, except for a boundary point, thereby providing a sharp characterization of consistent estimability of the slope for these processes. The results are extended to multivariate slopes and to accommodate measurement error using local-averaging-and-differencing-based estimators. We show that differencing-based estimation remains consistent for popular classes of nonstationary Gaussian random fields, some non-Gaussian random fields and irregular designs. The finite sample behavior of the estimators is explored via numerical experiments.