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
2025 年的期刊文章 · 第 16 页
共 1491 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2025-06-19 到 2025-06-05。
按期刊、按主题看 →
SMMR Latent Variables & Mixtures Clinical Trials Spatial & Time Series · 发表于 2025-06-19
Jerome Johnson, Xiangyu Yu, Suzanne M Dufault, Nicholas P Jewell
摘要
A recent large-scale cluster randomized test-negative study assessed the impact of a mosquito-based intervention on the incidence of clinical dengue showing a protective efficacy of 77.1% (95% CI: (65.3%, 84.9%)). While the intervention was randomized at a cluster-level, human and mosquito movement suggest potential violations in assumptions necessary for intention-to-treat analyses to produce accurate estimates of the full intervention effect due to spatial clustering of dengue cases, and/or potential non-independence in the intervention arising from spillover of the intervention (or control) across cluster boundaries. We address these distinct but related effects using two approaches. First, we examine whether a clustering effect exists, that is, whether the presence of a recent dengue case in the sample within a specified distance from a residence raises the risk of dengue. Second, we use cluster reallocation techniques to examine intervention spillover effects. We find strong spatial effects of the presence of dengue cases on the risk of clinical dengue that exhibit both serospecificity and a dose response, more evident in control than intervention clusters at least on an additive scale. Contrarily, there is no evidence of any appreciable local spillover effect from intervention to control clusters, or vice versa, in terms of either the risk of dengue infection or the level of disease clustering.
SMMR Causal Inference Missing Data & Measurement Error · 发表于 2025-06-19
Sevinc Puren Yucel Karakaya, Ilker Unal
摘要
One of the primary challenges encountered in propensity score (PS) weighting is the presence of observations with missing covariates. In such cases, several potential solutions based on multiple imputation have been proposed. The most prevalent of these is the MI te method, which combines treatment effect estimates derived from imputed datasets. A limited number of PS studies have incorporated the MI te method with the missing indicator method; however, these studies only incorporated the missing indicator into the PS model. The aim of this simulation study is to propose two novel methods that incorporate the missing indicator approach with the MI te . This incorporation either entails including the missing indicator into the outcome model (MIMI o ) or, alternatively, into both the outcome and PS model (MIMI pso ). The construction of the simulation scenarios was predicated on three elements: the mechanism of missing data, the type of treatment effect, and the presence of unmeasured confounding. In the presence of unmeasured confounding, the MIMI pso method was the most effective method under the MAR mechanism. In the context of the MNAR mechanism, the method that exhibited the lowest bias was MIMI o for homogeneous treatment effect and MIMI pso for heterogeneous treatment effect. The MI te method exhibited the highest levels of bias and variation. In view of the difficulties involved in identifying the mechanism of missing data, the variability in treatment effects across subgroups and the potential for unmeasured confounding variables in practice, researchers are encouraged to utilize the MIMI pso method.
SMMR Networks & Graphs Latent Variables & Mixtures Diagnostics & Prediction · 发表于 2025-06-19
Jiaxing Qiu, Douglas E Lake, Pavel Chernyavskiy, Teague R Henry
摘要
For prediction models developed on clustered data that do not account for cluster heterogeneity in model parameterization, it is crucial to use cluster-based validation to assess model generalizability on unseen clusters. This article introduces a clustered estimator of the network information criterion to approximate leave-one-cluster-out deviance for standard prediction models with twice-differentiable log-likelihood functions. The clustered network information criterion serves as a fast alternative to cluster-based cross-validation. Stone proved that the Akaike information criterion is asymptotically equivalent to leave-one-observation-out cross-validation for true parametric models with independent and identically distributed observations. Ripley noted that the network information criterion, derived from Stone’s proof, is a better approximation when the model is misspecified. For clustered data, we derived clustered network information criterion by substituting the Fisher information matrix in the network information criterion with a clustering-adjusted estimator. The clustered network information criterion imposes a greater penalty when the data exhibits stronger clustering, thereby allowing the clustered network information criterion to better prevent over-parameterization. In a simulation study and an empirical example, we used standard regression to develop prediction models for clustered data with Gaussian or binomial responses. Compared to the commonly used Akaike information criterion and Bayesian information criterion for standard regression, clustered network information criterion provides a much more accurate approximation to leave-one-cluster-out deviance and results in more accurate model size and variable selection, as determined by cluster-based cross-validation, especially when the data exhibit strong clustering.
SMMR Longitudinal Data Optimization & Learning Theory · 发表于 2025-06-19
Mitchell Aaron Schepps, Jérémy Seurat, France Mentré, Weng Kee Wong
摘要
Lithium is recommended as a first line treatment for patients with bipolar disorder. However, only certain patients show a good response to the drug, and the variability and tolerability of lithium response are poorly understood. Greater precision in the early identification of individuals who are likely to respond well to lithium is a significant unmet clinical need. We create optimal designs to better understand the pharmacokinetic exposition of lithium for patients with and without a genetic covariate. From a Fisher information matrix based method, we find different optimal designs for estimating various parameters in a complicated pharmacokinetics/pharmacodynamics nonlinear mixed effects model with multiple physician specified constraints. Our approach uses flexible state-of-the-art metaheuristics to find various types of efficient designs, including multiple-objective optimal designs that can balance the competitiveness of the objectives and deliver higher efficiencies for more important objectives. Results from this article will be used as part of a broader study to implement efficient designs to better understand the exposition of sustained-release lithium in patients with bipolar disorder.
SMMR Clinical Trials · 发表于 2025-06-18
Isabelle Wilson, Steven Julious, Christina Yap, Susan Todd, Munyaradzi Dimairo
摘要
Introduction: Adaptive designs (ADs) offer clinical trials flexibility to modify design aspects based on accumulating interim data. Response adaptive randomisation (RAR) adjusts treatment allocation according to interim results, favouring promising treatments. Despite scientific appeal, RAR adoption lags behind other ADs. Understanding methods and applications could provide insights and resources and reveal future research needs. This study examines RAR application, trial results and achieved benefits, reporting gaps, statistical tools and concerns, while highlighting examples of effective practices. Methods: RAR trials with comparative efficacy, effectiveness or safety objectives, classified at least phase I/II, were identified via statistical literature, trial registries, statistical resources and researcher-knowledge. Search spanned until October 2023, including results until February 2024. Analysis was descriptive and narrative. Results: From 652 articles/trials screened, 65 planned RAR trials (11 platform trials) were identified, beginning in 1985 and gradually increasing through to 2023. Most trials were in oncology (25%) and drug-treatments (80%), with 63% led by US teams. Predominantly Phase II (62%) and multi-arm (63%), 85% used Bayesian methods, testing superiority hypotheses (86%). Binary outcomes appeared in 55%, with a median observation of 56 days. Bayesian RAR algorithms were applied in 83%. However, 71% of all trials lacked clear details on statistical implementation. Subgroup-level RAR was seen in 23% of trials. Allocation was restricted in 51%, and 88% was included a burn-in period. Most trials (85%) planned RAR alongside other adaptations. Of trials with results, 92% used RAR, but over 50% inadequately reported allocation changes. A mean 22% reduction in sample size was seen, with none over-allocating to ineffective arms. Conclusion: RAR has shown benefits in conditions like sepsis, COVID-19 and cancer, enhancing effective treatment allocation and saving resources. However, complexity, costs and simulation need limit wider adoption. This review highlights RAR's benefits and suggests enhancing statistical tools to encourage wider adoption in clinical research.
Biometrika Causal Inference · 发表于 2025-06-17
Zehao Su, Helene C Rytgaard, Henrik Ravn, Frank Eriksson
摘要
Summary: We consider the problem of indirect comparison, where a treatment arm of interest is absent by design in one randomized controlled trial, but available in the other. The former is the target trial, and the latter is the source trial. The identifiability of the target population average treatment effect often relies on conditional transportability assumptions. However, it is a common concern whether all relevant effect modifiers are measured and controlled for. We give a new proximal identification result in the presence of shifted, unobserved effect modifiers based on proxies: an adjustment proxy in both trials and an additional reweighting proxy in the source trial. We propose an estimator that is doubly robust against misspecifications of the so-called bridge functions and asymptotically normal under mild consistency of estimators for the bridge functions. We use two weight management trials as a context to illustrate selection of proxies and apply our method to compare the weight loss effect of active treatments from these trials.
SMMR Clinical Trials Testing & Uncertainty Causal Inference · 发表于 2025-06-17
Lara Maleyeff, Fan Li, Sebastien Haneuse, Rui Wang
摘要
Cluster randomized trials are widely used in healthcare research for the evaluation of intervention strategies. Beyond estimating the average treatment effect, it is often of interest to assess whether the treatment effect varies across subgroups. While conventional methods based on tests of interaction terms between treatment and covariates can be used to detect treatment effect heterogeneity in cluster randomized trials, they typically rely on parametric assumptions that may not hold in practice. Adapting existing permutation tests from individually randomized trials, however, requires conceptual clarification and modification due to the multiple possible interpretations of treatment effect heterogeneity in the cluster randomized trial context. In this work, we develop variations of permutation tests and clarify key causal definitions in order to assess treatment effect heterogeneity in cluster randomized trials. Our procedure enables investigators to simultaneously test for effect modification across a large number of covariates, while maintaining nominal type I error rates and reasonable power in simulation studies. In the Pain Program for Active Coping and Training (PPACT) study, the proposed methods are able to detect treatment effect heterogeneity that was not identified by conventional methods assessing treatment–covariate interactions.
Biometrika Causal Inference Longitudinal Data Nonparametric & Semiparametric · 发表于 2025-06-17
Pan Zhao, Yifan Cui
摘要
Summary: Recently, there has been a surge in methodological development for the difference-in-differences approach to evaluate causal effects. Standard methods in the literature rely on the parallel-trend assumption to identify the average treatment effect on the treated. However, the parallel-trend assumption may be violated in the presence of unmeasured confounding, and the average treatment effect on the treated may not be useful in learning a treatment assignment policy for the entire population. In this article, we propose a general instrumented difference-in-differences approach for learning the optimal treatment policy. Specifically, we establish identification results using a binary instrumental variable when the parallel-trend assumption fails to hold. Additionally, we construct a Wald estimator, novel inverse probability weighting estimators and a class of semiparametric efficient and multiply robust estimators, with theoretical guarantees on consistency and asymptotic normality, even when relying on flexible machine learning algorithms for nuisance parameter estimation. Furthermore, we extend the instrumented difference in differences to the panel data setting. We evaluate our methods in extensive simulations and in an analysis of the Australian Longitudinal Survey.
SiM Clinical Trials · 发表于 2025-06-13
Nicoletta D'Angelo
摘要
This paper introduces a new method for change detection in medical and psychometric studies based on the recently introduced pseudo Score statistic, for which the sampling distribution under the alternative hypothesis has been determined. Our approach has the advantage of simplicity in its computation, eliminating the need for resampling or simulations to obtain critical values. Additionally, it comes with known null and alternative distributions, facilitating easy calculations for power levels and sample size planning. The paper indeed also discusses the topic of power analysis in segmented regression, namely the estimation of sample size or power level when the study data being collected focuses on a covariate expected to affect the mean response via a piecewise relationship with an unknown breakpoint. We run simulation studies showing that our method outperforms other Tests for a Change Point (TFCP) with both normally distributed and binary data and carry out two real data analyses on genomic data and SAT critical reading data. The proposed test contributes to the framework of medical and psychometric research, and it is available on the Comprehensive R Archive Network (CRAN) and in a more user‐friendly Shiny App, both illustrated at the end of the paper.
Biometrics Clinical Trials · 发表于 2025-06-13
Nina Deliu, Sofia S Villar
摘要
It is now commonly known that using response-adaptive designs for data collection offers great potential in terms of optimizing expected outcomes, but poses multiple challenges for inferential goals. In many settings, such as phase-II or confirmatory clinical trials, a main barrier to their practical use is the lack of type-I error guarantees and/or power efficiency, especially in finite samples. This work addresses this gap. Specifically, focusing on a novel test statistic defined on the randomization probabilities of the (randomized) adaptive design, we derive its finite-sample and asymptotic guarantees. Further theoretical properties are evaluated for Thompson sampling, a Bayesian response-adaptive design that is commonly used both in clinical applications and beyond (eg, recommendation systems or mobile health). The frequentist error control advantages of the proposed approach—also able to preserve expected outcome optimalities—are illustrated in a real-world phase-II oncology trial and in simulation experiments.
SiM Longitudinal Data Clinical Trials · 发表于 2025-06-13
Corine Baayen, Paul Blanche, Christopher Jennison, Brice Ozenne
摘要
Group sequential trials (GST) allow for early stopping of a clinical trial for efficacy or futility, without compromising its validity. Statistical methodology for GST is well established when the endpoint is observed immediately, but less so for endpoints that are measured with a delay, such as repeatedly measured outcomes for which the primary measurement of interest is taken after several months. The latter can result in pipeline subjects at an interim analysis. These subjects may have early outcome measurements available, but their final endpoint is yet to be observed. Accounting for these early measurements has been shown to increase statistical power. Most importantly, pipeline patients will contribute with additional data after a decision to stop enrollment has been taken at an interim analysis. To make the best use of all available data, these data are ideally incorporated in the final analysis in a formal way. In this tutorial paper, we provide guidance on how to plan a GST with repeated measurements and a delayed endpoint and how to analyze the data resulting from these trials. We discuss existing methods, and also expand on them, for example adding a nonbinding stopping rule for futility and working out the computational details to derive valid p‐values and confidence intervals. We also provide an R package and R code to perform the methods discussed in this paper.
SMMR Causal Inference Missing Data & Measurement Error Nonparametric & Semiparametric · 发表于 2025-06-12
Shaojie Wei, Qinpeng He, Wei Li, Zhi Geng
摘要
Numerous estimation procedures employed in causal inference often rely on accurately measured data. However, the prevalence of measurement errors in practical studies may yield biased effect estimates. It is common to employ validation samples to rectify such biases in the measurement error literature. This article focuses on the estimation of the average causal effect with a misclassified binary treatment in a primary population of interest. By leveraging a validation sample with covariates, an error-prone version of treatment and a true treatment recorded, we provide identifiability results under certain conditions. Building on identifiability, we explore three classes of estimators, each demonstrating consistency and asymptotic normality within distinct model sets. Furthermore, we propose a multiply robust estimation approach for the treatment effect based on the semiparametric theory framework. The multiply robust estimator retains consistent under any one of the listed model sets and achieves the semiparametric efficiency bound, provided all models are correct. We demonstrate the satisfactory performance of the proposed estimators through simulation studies and a real data analysis.
Biometrika Survival Analysis Causal Inference Diagnostics & Prediction · 发表于 2025-06-12
Hanxuan Ye, Hongzhe Li
摘要
Summary: Traditional statistical and machine learning methods typically assume that the training and test data follow the same distribution. However, this assumption is frequently violated in real-world applications, where the training data in the source domain may underrepresent specific subpopulations in the test data of the target domain. This article addresses target-independent learning under covariate shift, focusing on multicalibration for survival probability and restricted mean survival time. A black-box post-processing boosting algorithm specifically designed for censored survival data is introduced. By leveraging pseudo-observations, our method produces a multicalibrated predictor that is competitive with inverse propensity score weighting in predicting the survival outcome in an unlabelled target domain, ensuring, not only overall accuracy, but also fairness across diverse subpopulations. Our theoretical analysis of pseudo-observations builds upon the functional delta method and the $ p $-variational norm. The algorithm’s sample complexity, convergence properties and multicalibration guarantees for post-processed predictors are provided. The results establish a fundamental connection between multicalibration and universal adaptability, demonstrating that our calibrated function is comparable to, or outperforms, the inverse propensity score weighting estimator. Extensive numerical simulations and a real-world case study on cardiovascular disease risk prediction using two large prospective cohort studies validate the effectiveness of our approach.
SiM Clinical Trials · 发表于 2025-06-11
Ana M. Ortega‐Villa, Megan C. Grieco, Kevin Rubenstein, Jing Wang, Michael A. Proschan
摘要
At the beginning of a phase III clinical trial, there is great optimism. After all, the phase II trial results were encouraging. Then, early data from the phase III trial trend in the wrong way, but there is still an opportunity for the trend to reverse and become statistically significant by the end. At what point does optimism become denial of reality? How do we decide when a clinical trial is futile? What does futility even mean? This tutorial reviews different concepts and tools for evaluating futility, including conditional and predictive power, reverse conditional power, predicted interval plots, revised unconditional power, and beta spending functions.
SiM Diagnostics & Prediction Econometrics & Economics · 发表于 2025-06-11
Qiqi Deng, Qian Li, Naitee Ting, Feng Yu
摘要
With the advance of medical sciences and better understanding of human biological systems, the next generation of treatment has shifted toward personalized medicine. It is expected that personalized medicine, such as molecularly targeted anti‐cancer agents, is more efficacious in marker‐positive patients, while marker‐negative patients may or may not benefit from the treatment. Due to technology limitations in marker identification and incomplete understanding of the role of biomarkers in treatment effect, it is possible that the marker is not predictive. Therefore, it is often of interest to test the treatment on the overall population as well as the biomarker‐positive subgroup. Testing both the overall population and the biomarker‐positive subgroup introduces a multiplicity issue and leads to type I error inflation if not adjusted appropriately. The available multiplicity adjustment procedures may not consider the logic needed in the two tests. A new method is proposed by applying the logical connections between the two hypothesis tests, and arbitrages between different rejection regions to make the testing strategy not only powerful but also sensible.
SiM Survival Analysis Diagnostics & Prediction · 发表于 2025-06-10
Peter C. Austin, Daniele Giardiello
摘要
The Cox proportional hazards regression model is frequently used to develop clinical prediction models for time‐to‐event outcomes, allowing clinicians to estimate an individual's risk of experiencing the outcome within specified time horizons (e.g., estimate an individual's 10‐year risk of death). The Cox regression model models the association between covariates and the hazard of the outcome. A key assumption of the Cox model is the proportional hazards assumption: the ratio of the hazard function for any two individuals is constant over time, and the ratio is a function of only their covariates and the regression coefficients. Calibration is an important aspect of the validation of clinical prediction models. Calibration refers to the concordance between predicted and observed risk. The impact of the violation of the proportional hazards assumption on the calibration of clinical prediction models developed using the Cox model has not been examined. We conducted a set of Monte Carlo simulations to assess the impact of the magnitude of the violation of the proportional hazards assumption on the calibration of the Cox model. We compared the calibration of predictions obtained using a Cox regression model that ignored the violation of the proportional hazards assumption with those obtained using accelerated failure time (AFT) models, Royston and Parmar's spline‐based parametric survival models, and generalized linear models using pseudo‐observations. We found that violation of the proportional hazards assumption had negligible impact on the calibration of predictions obtained using a Cox model.
SiM Other · 发表于 2025-06-10
Ethan Ashby, Bo Zhang, Genevieve G. Fouda, Youyi Fong, Holly Janes
摘要
Adjustment for prognostic baseline variables can reduce bias due to covariate imbalance and increase efficiency in randomized trials. While the use of covariate adjustment in late‐phase trials is justified by favorable large‐sample properties, it is seldom used in small, early‐phase studies, due to uncertainty in which variables are prognostic and the potential for precision loss, type I error rate inflation, and undercoverage of confidence intervals. To address this problem, we consider adjustment for a valid negative control outcome (NCO), or an auxiliary post‐randomization outcome believed to be completely unaffected by treatment but more highly correlated with the primary outcome than baseline covariates. We articulate the assumptions that permit adjustment for NCOs without producing post‐randomization selection bias, and describe plausible data‐generating models where NCO adjustment can improve upon adjustment for baseline covariates alone. In numerical experiments, we illustrate performance and provide practical recommendations regarding model selection and finite‐sample variance corrections. We apply our methods to the reanalysis of two early‐phase vaccine trials in HIV exposed uninfected (HEU) infants, where we demonstrate that adjustment for auxiliary post‐baseline immunological parameters can enhance the precision of vaccine effect estimates relative to standard approaches that avoid adjustment or adjust for baseline covariates alone.
SiM Missing Data & Measurement Error Testing & Uncertainty · 发表于 2025-06-10
Chung‐Han Lee
摘要
Confidence interval estimation for a risk difference is commonly used in various applications. The method of variance estimates recovery (MOVER) is a useful method for constructing the confidence interval of the risk difference. The confidence interval estimation with incomplete data has been widely studied in recent years, as missing values can occur during data collection. In this study, for the Poisson and binomial distributions, we propose proper multiple imputation procedures for the MOVER to estimate the confidence intervals for the risk difference, not only for missing at random but also for missing not at random. A simulation study shows that the coverage probabilities of the proposed intervals are closer to the nominal level than those of existing intervals, particularly when the true parameters are near the boundaries. These multiple imputation confidence intervals are illustrated with real data examples.
SiM Bayesian Methods Nonparametric & Semiparametric Survival Analysis · 发表于 2025-06-10
Chunjie Wang, Zhexin Lu, Chuchu Wang, Xinyuan Song
摘要
The information extracted from imaging data has become increasingly important in disease diagnosis as it uncovers associations between imaging features and diseases of interest. This study proposes a partial functional Tobit censored quantile regression (PFTCQR) model to investigate the quantile‐specific relationships between the time of incidence of laryngeal cancer and a set of imaging and clinical predictors based on the data collected from a laryngeal cancer study in the Otolaryngology Department of a tertiary hospital in Jilin Province, China. The functional principal component analysis and moment method are employed to estimate the slope and covariance functions of the functional predictors. An efficient Markov chain Monte Carlo (MCMC) algorithm is developed, leveraging the location‐scale mixture representation of the asymmetric Laplace distribution (ALD) to perform the estimation. Furthermore, we extend the PFTCQR model to the composite quantile regression framework and incorporate variable selection for scalar covariates, further enhancing the robustness and efficiency of parameter estimation and improving model fitting. The proposed method is demonstrated through simulation studies and applied to the laryngeal carcinoma data. Results provide new insights into potential risk factors for laryngeal carcinoma and their effects varying across quantiles. Specific laryngeal regions are identified as significantly associated with the progression of the disease.
SMMR Causal Inference Functional Data · 发表于 2025-06-09
Kaidi Kong, Li Guan, Zhongzhan Zhang
摘要
In personalized medicine study, how to estimate the optimal individualized treatment regime based on available individual information is a fundamental problem. In recent years, functional data analysis has appeared extensively in medical research, while the optimal individualized treatment regime based on the combination of scalar covariates and functional covariates have rarely been studied and the only few studies are mostly conducted in the context of randomized trials. In this article, we propose a flexible regression-based approach in which the outcome variable is real-valued and the covariates contain multiple scalar covariates and a functional covariate. Our approach is applicable to both randomized trials and observational studies, and the convergence rates of the proposed optimal individualized treatment regime estimators are presented for both situations. Sufficient simulation studies and a real data analysis are conducted to justified the validity of our proposed method.
JASA Bayesian Methods · 发表于 2025-06-09
David B. Dahl, Richard L. Warr, Thomas P. Jensen
摘要
Although exchangeable processes from Bayesian nonparametrics have been used as a generating mechanism for random partition models, we deviate from this paradigm to explicitly incorporate clustering information in the formulation of our random partition model. Our shrinkage partition distribution takes any partition distribution and shrinks its probability mass toward a specific anchor partition. We show how this provides a framework to model hierarchically-dependent and temporally-dependent random partitions. The shrinkage parameter controls the degree of dependence, accommodating at its extremes both independence and complete equality. Since prior knowledge of items may vary, our formulation allows the degree of shrinkage toward the anchor to be item-specific. Our random partition model has a tractable normalizing constant which allows for standard Markov chain Monte Carlo algorithms for posterior sampling. We prove intuitive theoretical properties for our distribution and compare it to related partition distributions. We show that our model provides better out-of-sample fit in a real data application. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*
SiM Causal Inference Clinical Trials Nonparametric & Semiparametric · 发表于 2025-06-07
Wenli Liu, Jing Qin, Yukun Liu
摘要
Randomized controlled trials (RCTs) are the gold standard for causal inference and are widely used. However, valid analyses of RCTs are often complicated by non‐compliance, which can lead to confounding bias and biased causal effect estimation. The main challenge comes from compliance datasets in both the treatment and control groups both following two‐component mixture models. The maximum nonparametric likelihood estimator is inconsistent in a two‐component mixture model even if the mixture proportion and one of the components are completely known, but the other component is unknown. In this paper, we instead assume parametric models for the ratios of risks among compliers assigned treatment, never‐takers and always‐takers, and leave the baseline compliers not assigned treatment unspecified. We develop a novel two‐step maximum likelihood estimation procedure by making full use of the observed covariates and latent compliance classes, which theoretically can produce asymptotic root consistent estimators. In particular, our proposed estimator for the conditional local risk ratio always lies within the range of the parameter. Our numerical results show that the proposed method is generally more reliable than existing alternatives.
SiM Clinical Trials Survival Analysis · 发表于 2025-06-07
Pei‐Fang Su, Chieh‐Chi Wu
摘要
Estimating the required sample size is an essential task for a clinical trial. It ensures statistical power and makes it possible to draw meaningful conclusions from the study results. Moreover, to minimize treatment imbalances within each covariate subgroup, covariate‐adaptive randomization is a popular method. The first aim of this study is to investigate the required sample size for covariate‐adaptive randomization based on survival outcomes. We evaluate the testing performance using the calculated sample size under simple randomization, stratified permuted block randomization, and covariate‐adaptive biased coin randomization. To provide preliminary insights into the trial's progress and potential efficacy, an interim analysis is commonly conducted. The second aim of the study is to provide a strategy for interim analysis in covariate‐adaptive randomization trials, which involves stopping the trial early based on accumulating data if one treatment arm proves to be significantly more effective or harmful than the others. Because the data collected before and after the interim monitoring time are dependent, performing multiple tests may inflate the overall Type I error rate. We thus re‐estimate the required sample size, propose a valid hypothesis testing method for interim analysis, and study the underlying theoretical properties of the testing statistics, incorporating covariate‐adaptive randomization. The performance of the proposed formula and interim strategy is evaluated through comprehensive simulations, including a sensitivity analysis. We also provide the R code to benefit the readers. Finally, two survival trials, the RE01 trial and a lung cancer study, are treated as pilot studies, and we show that the proposed strategies are valid under covariate‐adaptive randomization.
Biometrika Causal Inference Networks & Graphs · 发表于 2025-06-06
Juraj Bodik, Valérie Chavez-Demoulin
摘要
Summary: We consider the problem of learning a set of direct causes of a target variable from an observational joint distribution. Learning directed acyclic graphs that represent the causal structure is a fundamental problem in science. Several results are known when the full directed acyclic graph is identifiable from the distribution, such as when a nonlinear Gaussian data-generating process is assumed. Here, we are interested only in identifying the direct causes of one target variable (local causal structure), not the full directed acyclic graph. This allows us to relax the identifiability assumptions and develop possibly faster and more robust algorithms. In contrast to the invariance causal prediction framework, here the only assumption is that we observe one environment without any interventions. We discuss different assumptions for the data-generating process of the target variable under which the set of direct causes is identifiable from the distribution. While doing so, we impose essentially no assumptions on the variables other than the target variable. In addition to the novel identifiability results, we provide two practical algorithms for estimating the direct causes from a finite random sample and demonstrate their effectiveness on several benchmark and real datasets.
JASA Causal Inference Testing & Uncertainty High-Dimensional · 发表于 2025-06-06
Jin-Hong Du, Larry Wasserman, Kathryn Roeder
摘要
Tens of thousands of simultaneous hypothesis tests are routinely performed in genomic studies to identify differentially expressed genes. However, due to unmeasured confounders, many standard statistical approaches may be substantially biased. This paper investigates the large-scale hypothesis testing problem for multivariate generalized linear models in the presence of confounding effects. Under arbitrary confounding mechanisms, we propose a unified statistical estimation and inference framework that harnesses orthogonal structures and integrates linear projections into three key stages. It begins by disentangling marginal and uncorrelated confounding effects to recover the latent coefficients. Subsequently, latent factors and primary effects are jointly estimated through lasso-type optimization. Finally, we incorporate projected and weighted bias-correction steps for hypothesis testing. Theoretically, we establish the identification conditions of various effects and non-asymptotic error bounds. We show effective Type-I error control of asymptotic z -tests as sample and response sizes approach infinity. Numerical experiments demonstrate that the proposed method controls the false discovery rate by the Benjamini-Hochberg procedure and is more powerful than alternative methods. By comparing single-cell RNA-seq counts from two groups of samples, we demonstrate the suitability of adjusting confounding effects when significant covariates are absent from the model. *(abstract via PubMed)*
JASA Networks & Graphs · 发表于 2025-06-06
Ting Yan, Yuanzhang Li, Jinfeng Xu, Yaning Yang, Ji Zhu
摘要
We explore the Wilks phenomena in two random graph models: the β-model and the Bradley–Terry model. For two increasing dimensional null hypotheses, including a specified null H0:βi=βi0 for i=1,…,r and a homogenous null H0:β1=⋯=βr, we reveal high dimensional Wilks’ phenomena that the normalized log-likelihood ratio statistic, [2{l(β̂)−l(β̂0)}−r]/(2r)1/2, converges in distribution to the standard normal distribution as r goes to infinity. Here, l(β) is the log-likelihood function on the model parameter β=(β1,…,βn)⊤, β̂ is its maximum likelihood estimator (MLE) under the full parameter space, and β̂0 is the restricted MLE under the null parameter space. For the homogenous null with a fixed r, we establish Wilks-type theorems that 2{l(β̂)−l(β̂0)} converges in distribution to a chi-square distribution with r−1 degrees of freedom, as the total number of parameters, n, goes to infinity. When testing the fixed dimensional specified null, we find that its asymptotic null distribution is a chi-square distribution in the β-model. However, unexpectedly, this is not true in the Bradley–Terry model. By developing several novel technical methods for asymptotic expansion, we explore Wilks-type results in a principled manner; these principled methods should be applicable to a class of random graph models beyond the β-model and the Bradley–Terry model. Simulation studies and real network data applications further demonstrate the theoretical results. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*
JASA Causal Inference Genetics & Omics Testing & Uncertainty · 发表于 2025-06-06
Jin-Hong Du, Zhenghao Zeng, Edward H. Kennedy, Larry Wasserman, Kathryn Roeder
摘要
With the evolution of single-cell RNA sequencing techniques into a standard approach in genomics, it has become possible to conduct cohort-level causal inferences based on single-cell-level measurements. However, the individual gene expression levels of interest are not directly observable; instead, only repeated proxy measurements from each individual's cells are available, providing a derived outcome to estimate the underlying outcome for each of many genes. In this paper, we propose a generic semiparametric inference framework for doubly robust estimation with multiple derived outcomes, which also encompasses the usual setting of multiple outcomes when the response of each unit is available. To reliably quantify the causal effects of heterogeneous outcomes, we specialize the analysis to standardized average treatment effects and quantile treatment effects. Through this, we demonstrate the use of the semiparametric inferential results for doubly robust estimators derived from both Von Mises expansions and estimating equations. A multiple testing procedure based on Gaussian multiplier bootstrap is tailored for doubly robust estimators to control the false discovery exceedance rate. Applications in single-cell CRISPR perturbation analysis and individual-level differential expression analysis demonstrate the utility of the proposed methods and offer insights into the usage of different estimands for causal inference in genomics. *(abstract via PubMed)*
JASA Other · 发表于 2025-06-06
Yiling Xie, Xiaoming Huo
摘要
Adversarial training has been proposed to protect machine learning models against adversarial attacks. This paper focuses on adversarial training under 𝓁∞-perturbation, which has recently attracted much research attention. The asymptotic behavior of the adversarial training estimator is investigated in the generalized linear model. The results imply that the asymptotic distribution of the adversarial training estimator under 𝓁∞-perturbation could put a positive probability mass at 0 when the true parameter is 0, providing a theoretical guarantee of the associated sparsity-recovery ability. Alternatively, a two-step procedure is proposed—adaptive adversarial training, which could further improve the performance of adversarial training under 𝓁∞-perturbation. Specifically, the proposed procedure could achieve asymptotic variable-selection consistency and unbiasedness. Numerical experiments are conducted to show the sparsity-recovery ability of adversarial training under 𝓁∞-perturbation and to compare the empirical performance between classic adversarial training and adaptive adversarial training. *(abstract via OpenAlex)*
SiM Other · 发表于 2025-06-06
Mingchi Xu, Dirk Douwes‐Schultz, Alexandra M. Schmidt
摘要
In epidemiological studies, zero‐inflated and hurdle models are commonly used to handle excess zeros in reported infectious disease cases. However, they cannot model the persistence (transition from presence to presence) and reemergence (transition from absence to presence) of a disease separately. Covariates can sometimes have different effects on the reemergence and persistence of a disease. Recently, a zero‐inflated Markov switching negative binomial model was proposed to accommodate this issue. We introduce a Markov switching negative binomial hurdle model as a competitor of that approach, as hurdle models are often also used as alternatives to zero‐inflated models for accommodating excess zeroes. We begin the comparison by inspecting the underlying assumptions made by both models. Hurdle models assume perfect detection of the disease cases while zero‐inflated models implicitly assume the case counts can be under‐reported, thus, we investigate when a negative binomial distribution can approximate the true distribution of reported counts. A comparison of the fit of the two types of Markov switching models is undertaken on chikungunya cases across the neighborhoods of Rio de Janeiro. We find that, among the fitted models, the Markov switching negative binomial zero‐inflated model produces the best predictions, and both Markov switching models produce remarkably better predictions than more traditional negative binomial hurdle and zero‐inflated models.
SiM Bayesian Methods Causal Inference Longitudinal Data · 发表于 2025-06-06
Ruyi Liu, Liangyuan Hu, Francis Perry Wilson, Joshua L. Warren, Fan Li
摘要
In longitudinal observational studies with time‐varying confounders, the generalized computation algorithm formula (g‐formula) is a principled tool to estimate the average causal effect of a treatment regimen. However, the standard non‐iterative g‐formula implementation requires specifying both the conditional distribution of the outcomes and the joint distribution of all time‐varying covariates. This process can be cumbersome to implement and is prone to model misspecification bias. As an alternative, the iterative conditional expectation (ICE) g‐formula estimator solely depends on a series of nested outcome regressions and avoids the need for specifying the full distribution of all time‐varying covariates. This simplicity lends itself to the natural integration of flexible machine learning techniques to develop more robust average causal effect estimators with time‐varying treatments. In this work, we introduce a Bayesian approach that includes parametric regressions and Bayesian Additive Regression Trees to flexibly model a series of outcome surfaces. We fit the ICE g‐formula and develop a sampling algorithm to obtain samples from the posterior distribution of the final causal effect estimator. We illustrate the performance characteristics of the Bayesian ICE estimator and the associated variations via simulation studies and applications to two real world data examples.
SiM Genetics & Omics Survival Analysis · 发表于 2025-06-05
Chin‐Chun Chen, Sheng‐Mao Chang, Peng‐Chan Lin, Pei‐Fang Su
摘要
In the form of multidimensional arrays, modeling considering tensor covariates is capable of capturing important feature interactions or encompassing a broader range of data structures. In the study, we proposed a statistical inference based on tensor‐based covariates for the Cox regression model in the integrated genome study. The proposed method stands out for reducing the parameter dimension within the partial likelihood function through the application of rank‐R decomposition. We introduced the method of estimation and rank selection. Subsequently, variable‐specific tensor tests are proposed to evaluate the significance of the covariance effect. Asymptotic normality is obtained for all the regression estimators. Simulation studies, including rank selection, estimation performance, and testing performance, demonstrate the effectiveness of the proposed algorithm. Overall, the proposed method provides readers with the necessary information to understand the genome effect of tensor‐based covariates. To benefit readers, the R code for implementing the method is provided. Lastly, we introduce the proposed method using a colorectal cancer dataset, addressing signal detection by simultaneously gathering information across different kinds of omics platforms. This dataset integrates clinical information with multi‐omics data, including copy number variation, methylation, and mRNA sequencing data, as tensor‐based covariates, and evaluates relationships involving right‐censored data.
SiM Machine Learning Networks & Graphs Longitudinal Data · 发表于 2025-06-05
Yimo Zhang, Jon A. Steingrimsson, Akhil Ambekar, Hwamee Oh, Ani Eloyan, Alzheimer's Disease Neuroimaging Initiative
摘要
Positron emission tomography (PET) imaging technology is widely used for diagnosing Alzheimer's disease (AD) in people with dementia. Although various computational methods have been proposed for diagnosis of AD using PET images, prediction of disease diagnosis by leveraging longitudinal PET imaging data has not been widely studied. In this paper, we propose novel implementations of deep learning models graph neural network (GNN) and transformer encoder (TE), to leverage longitudinal sequences of PET images in addition to cognitive scores for prediction of AD diagnosis and prediction of conversion from cognitively unimpaired or mild cognitive impairment to AD using data collected in the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. In addition, we compare the performance of these two approaches with other methods including simple feed‐forward networks (FFN) that only utilize PET images from a single time point and recurrent neural networks (RNN) that model longitudinal images by considering them as sequences, while failing to take into account between visit time variability. We show that GNN and TE have higher predictive performance than FFN and RNN in predicting disease diagnosis, where predictive performance is measured by accuracy, area under receiver operating characteristic (AUC) curve, and the Brier score. Furthermore, we illustrate the potential of PET images in predicting conversion from mild cognitive impairment or cognitively normal to AD.
SiM Genetics & Omics High-Dimensional · 发表于 2025-06-05
Matteo D'Alessandro, Theophilus Quachie Asenso, Manuela Zucknick
摘要
Modeling with multiomics data presents multiple challenges, such as the high dimensionality of the problem (), the presence of interactions between features, and the need for integration between multiple data sources. We establish an interaction model that allows for the inclusion of multiple sources of data from the integration of two existing methods, pliable lasso and cooperative learning. The integrated model is tested both on simulation studies and on real multiomics datasets for predicting labor onset and cancer treatment response. The results show that the model is effective in modeling multisource data in various scenarios where interactions are present, both in terms of prediction performance and selection of relevant variables.
SiM High-Dimensional Functional Data · 发表于 2025-06-05
Xiong Cai, Jiguo Cao, Xingyu Yan, Peng Zhao
摘要
We propose a new class of high‐dimensional multiresponse partially functional linear regressions (MR‐PFLRs) to investigate the relationship between scalar responses and a set of explanatory variables, which include both functional and scalar types. In this framework, both the dimensionality of the responses and the number of scalar covariates can diverge to infinity. To account for within‐subject correlation, we develop a functional principal component analysis (FPCA)‐based penalized weighted least squares estimation procedure. In this approach, the precision matrix is estimated using penalized likelihoods, and the regression coefficients are then estimated through the penalized weighted least squares method, with the precision matrix serving as the weight. This method allows for the simultaneous estimation of both functional and scalar regression coefficients, as well as the precision matrix, while identifying significant features. Under mild conditions, we establish the consistency, rates of convergence, and oracle properties of the proposed estimators. Simulation studies demonstrate the finite‐sample performance of our estimation method. Additionally, the practical utility of the MR‐PFLR model is showcased through an application to Alzheimer's disease neuroimaging initiative (ADNI) data.
SiM Bayesian Methods Clinical Trials · 发表于 2025-06-05
Jike Huang, Fan Jia, Jiaxuan Li, Wanqiu Xie, Zhiwei Rong, Lan Mi, Yuqin Song, Yan Hou
摘要
Integrating external data into a clinical trial can introduce systematic bias in estimates and inflate the study's type I error due to differences in study design and enrollment criteria. Existing prior designs for information borrowing lack the ability to dynamically adjust the weight based on the similarity between concurrent and external data. To address this challenge, we thereby introduce a novel method called the elastic commensurate prior (ECP), which combines the commensurate prior with the elastic prior method. By dynamically adjusting the weight of external data using a measure of congruence, this method demonstrates strong performance in maintaining power while providing adequate type I error control across different scenarios, including congruence, approximate congruence, and incongruence between external and concurrent data. Compared to existing methods such as the modified power prior, meta‐analytic‐predictive (MAP) prior, robust MAP prior, non‐informative prior, and fully informative prior, the ECP method is flexible and performs well across all settings. Furthermore, our method also allows for the integration of covariates in estimating data congruence for dynamic information borrowing, achieving both strong performance in power and adequate control of type I error. Overall, the ECP represents a promising option for leveraging external data in clinical trials, reducing costs by decreasing the sample size requirement, and thereby accelerating research and drug development timelines.