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 27 页
共 1491 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2025-02-01 到 2025-01-24。
按期刊、按主题看 →
AoS Nonparametric & Semiparametric · 发表于 2025-02-01
John H. J. Einmahl, Andrea Krajina, Juan Juan Cai
摘要
Multivariate regular variation is a common assumption in the statistics literature and needs to be verified in real-data applications. We develop a novel hypothesis test for multivariate regular variation, employing localized empirical likelihood. We establish the weak convergence of the test statistic to a nonstandard, distribution-free limit and hence can provide universal critical values for the test. We show the very good finite-sample behavior of the procedure through simulations and apply the test to several real-data examples.
AoS Other · 发表于 2025-02-01
Yuchen Zhou, Yuxin Chen
摘要
This paper is concerned with estimating the column subspace of a low-rank matrix X⋆∈Rn1×n2 from contaminated data. How to obtain optimal statistical accuracy while accommodating the widest range of signal-to-noise ratios (SNRs) becomes particularly challenging in the presence of heteroskedastic noise and unbalanced dimensionality (i.e., n2≫n1). While the state-of-the-art algorithm HeteroPCA emerges as a powerful solution for solving this problem, it suffers from “the curse of ill-conditioning,” namely, its performance degrades as the condition number of X⋆ grows. In order to overcome this critical issue without compromising the range of allowable SNRs, we propose a novel algorithm, called Deflated−HeteroPCA, that achieves near-optimal and condition-number-free theoretical guarantees in terms of both ℓ2 and ℓ2,∞ statistical accuracy. The proposed algorithm divides the spectrum of X⋆ into well-conditioned and mutually well-separated subblocks, and applies HeteroPCA to conquer each subblock successively. Further, an application of our algorithm and theory to two canonical examples—the factor model and tensor PCA—leads to remarkable improvement for each application.
AoS Bayesian Methods · 发表于 2025-02-01
Sirio Legramanti, Daniele Durante, Pierre Alquier
摘要
There has been an increasing interest on summary-free solutions for approximate Bayesian computation (abc) that replace distances among summaries with discrepancies between the empirical distributions of the observed data and the synthetic samples generated under the proposed parameter values. The success of these strategies has motivated theoretical studies on the limiting properties of the induced posteriors. However, there is still the lack of a theoretical framework for summary-free abc that (i) is unified, instead of discrepancy-specific, (ii) does not necessarily require to constrain the analysis to data generating processes and statistical models meeting specific regularity conditions, but rather facilitates the derivation of limiting properties that hold uniformly, and (iii) relies on verifiable assumptions that provide more explicit concentration bounds clarifying which factors govern the limiting behavior of the abc posterior. We address this gap via a novel theoretical framework that introduces the concept of Rademacher complexity in the analysis of the limiting properties for discrepancy-based abc posteriors, including in non-i.i.d. and misspecified settings. This yields a unified theory that relies on constructive arguments and provides more informative asymptotic results and uniform concentration bounds, even in those settings not covered by current studies. These key advancements are obtained by relating the asymptotic properties of summary-free abc posteriors to the behavior of the Rademacher complexity associated with the chosen discrepancy within the family of integral probability semimetrics (ips). The ips class extends summary-based distances, and also includes the widely implemented Wasserstein distance and maximum mean discrepancy (mmd), among others. As clarified in specialized theoretical analyses of popular ips discrepancies and via illustrative simulations, this new perspective improves the understanding of summary-free abc.
AoS High-Dimensional Optimization & Learning Theory · 发表于 2025-02-01
Yinan Shen, Jingyang Li, Jian-Feng Cai, Dong Xia
摘要
High-dimensional linear regression under heavy-tailed noise or outlier corruption is challenging, both computationally and statistically. Convex approaches have been proven statistically optimal but suffer from high computational costs, especially since the robust loss functions are usually nonsmooth. More recently, computationally fast nonconvex approaches via subgradient descent are proposed, which, unfortunately, fail to deliver a statistically consistent estimator even under sub-Gaussian noise. In this paper, we introduce a projected subgradient descent algorithm for both the sparse linear regression and low-rank linear regression problems. The algorithm is not only computationally efficient with linear convergence but also statistically optimal, be the noise Gaussian or heavy-tailed with a finite 1+ε moment. The convergence theory is established for a general framework and its specific applications to absolute loss, Huber loss and quantile loss are investigated. Compared with existing nonconvex methods, ours reveals a surprising phenomenon of two-phase convergence. In phase one, the algorithm behaves as in typical nonsmooth optimization that requires gradually decaying stepsizes. However, phase one only delivers a statistically suboptimal estimator, which is already observed in the existing literature. Interestingly, during phase two, the algorithm converges linearly as if minimizing a smooth and strongly convex objective function, and thus a constant stepsize suffices. Underlying the phase-two convergence is the smoothing effect of random noise to the nonsmooth robust losses in an area close but not too close to the truth. Numerical simulations confirm our theoretical discovery and showcase the superiority of our algorithm over prior methods.
AoS Bayesian Methods Nonparametric & Semiparametric · 发表于 2025-02-01
Richard Nickl, Grigorios A. Pavliotis, Kolyan Ray
摘要
We consider nonparametric statistical inference on a periodic interaction potential W from noisy discrete space-time measurements of solutions ρ=ρW of the nonlinear McKean–Vlasov equation, describing the probability density of the mean field limit of an interacting particle system. We show how Gaussian process priors assigned to W give rise to posterior mean estimators that exhibit fast convergence rates for the implied estimated densities ρ¯ towards ρW. We further show that if the initial condition ϕ is not too smooth and satisfies a standard deconvolvability condition, then one can consistently infer Sobolev-regular potentials W at convergence rates N−θ for appropriate θ>0, where N is the number of measurements. The exponent θ can be taken to approach 1/2 as the regularity of W increases corresponding to ‘near-parametric’ models.
AoS Machine Learning · 发表于 2025-02-01
Xiang Li, Feng Ruan, Huiyuan Wang, Qi Long, Weijie J. Su
摘要
Since ChatGPT was introduced in November 2022, embedding (nearly) unnoticeable statistical signals into text generated by large language models (LLMs), also known as watermarking, has been used as a principled approach to provable detection of LLM-generated text from its human-written counterpart. In this paper, we introduce a general and flexible framework for reasoning about the statistical efficiency of watermarks and designing powerful detection rules. Inspired by the hypothesis testing formulation of watermark detection, our framework starts by selecting a pivotal statistic of the text and a secret key—provided by the LLM to the verifier—to control the false positive rate (the error of mistakenly detecting human-written text as LLM-generated). Next, this framework allows one to evaluate the power of watermark detection rules by obtaining a closed-form expression of the asymptotic false negative rate (the error of incorrectly classifying LLM-generated text as human-written). Our framework further reduces the problem of determining the optimal detection rule to solving a minimax optimization program. We apply this framework to two representative watermarks—one of which has been internally implemented at OpenAI—and obtain several findings that can be instrumental in guiding the practice of implementing watermarks. In particular, we derive optimal detection rules for these watermarks under our framework. These theoretically derived detection rules are demonstrated to be competitive and sometimes enjoy a higher power than existing detection approaches through numerical experiments.
JASA Optimization & Learning Theory · 发表于 2025-01-31
Giuseppe Alfonzetti, Ruggero Bellio, Yunxiao Chen, Irini Moustaki
摘要
A composite likelihood is an inference function derived by multiplying a set of likelihood components. This approach provides a flexible framework for drawing inferences when the likelihood function of a statistical model is computationally intractable. While composite likelihood has computational advantages, it can still be demanding when dealing with numerous likelihood components and a large sample size. This paper tackles this challenge by employing an approximation of the conventional composite likelihood estimator based on a stochastic optimization procedure. This novel estimator is shown to be asymptotically normally distributed around the true parameter. In particular, based on the relative divergent rate of the sample size and the number of iterations of the optimization, the variance of the limiting distribution is shown to compound for two sources of uncertainty: the sampling variability of the data and the optimization noise, with the latter depending on the sampling distribution used to construct the stochastic gradients. The advantages of the proposed framework are illustrated through simulation studies on two working examples: an Ising model for binary data and a gamma frailty model for count data. Finally, a real-data application is presented, showing its effectiveness in a large-scale mental health survey. *(abstract via OpenAlex)*
SMMR Survival Analysis Causal Inference · 发表于 2025-01-31
Dane Isenberg, Michael O Harhay, Nandita Mitra, Fan Li
摘要
Patient-centered outcomes, such as quality of life and length of hospital stay, are the focus in a wide array of clinical studies. However, participants in randomized trials for elderly or critically and severely ill patient populations may have truncated or undefined non-mortality outcomes if they do not survive through the measurement time point. To address truncation by death, the survivor average causal effect has been proposed as a causally interpretable subgroup treatment effect defined under the principal stratification framework. However, the majority of methods for estimating the survivor average causal effect have been developed in the context of individually randomized trials. Only limited discussions have been centered around cluster-randomized trials, where methods typically involve strong distributional assumptions for outcome modeling. In this article, we propose two weighting methods to estimate the survivor average causal effect in cluster-randomized trials that obviate the need for potentially complicated outcome distribution modeling. We establish the requisite assumptions that address latent clustering effects to enable point identification of the survivor average causal effect, and we provide computationally efficient asymptotic variance estimators for each weighting estimator. In simulations, we evaluate our weighting estimators, demonstrating their finite-sample operating characteristics and robustness to certain departures from the identification assumptions. We illustrate our methods using data from a cluster-randomized trial to assess the impact of a sedation protocol on mechanical ventilation among children with acute respiratory failure.
JASA Optimization & Learning Theory High-Dimensional · 发表于 2025-01-31
Xinmeng Huang, Kan Xu, Donghwan Lee, Hamed Hassani, Hamsa Bastani, Edgar Dobriban
摘要
Large and complex datasets are often collected from several, possibly heterogeneous sources. Multitask learning methods improve efficiency by leveraging commonalities across datasets while accounting for possible differences among them. Here, we study multitask linear regression and contextual bandits under sparse heterogeneity, where the source/task-associated parameters are equal to a global parameter plus a sparse task-specific term. We propose a novel two-stage estimator called MOLAR that leverages this structure by first constructing a covariate-wise weighted median of the task-wise linear regression estimates and then shrinking the task-wise estimates toward the weighted median. Compared to task-wise least squares estimates, MOLAR improves the dependence of the estimation error on the data dimension. Extensions of MOLAR to generalized linear models and constructing confidence intervals are discussed in the article. We then apply MOLAR to develop methods for sparsely heterogeneous multitask contextual bandits, obtaining improved regret guarantees over single-task bandit methods. We further show that our methods are minimax optimal by providing a number of lower bounds. Finally, we support the efficiency of our methods by performing experiments on both synthetic data and the PISA dataset on student educational outcomes from heterogeneous countries. 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 Bayesian Methods · 发表于 2025-01-31
Peng Wu, Shanshan Luo, Zhi Geng
摘要
There is growing interest in exploring causal effects in target populations via data combination. However, most approaches are tailored to specific settings and lack comprehensive comparative analyses. In this article, we focus on a typical scenario involving a source dataset and a target dataset. We first design six settings under covariate shift and conduct a comparative analysis by deriving the semiparametric efficiency bounds for the ATE in the target population. We then extend this analysis to six new settings that incorporate both covariate shift and posterior drift. Our study uncovers the key factors that influence efficiency gains and the “effective sample size” when combining two datasets, with a particular emphasis on the roles of the variance ratio of potential outcomes between two datasets and the derivatives of the posterior drift function. To the best of our knowledge, this is the first paper that explicitly explores the role of the posterior drift functions in causal inference. Additionally, we also propose novel methods for conducting sensitivity analysis to address violations of transportability between two datasets. We empirically validate our findings by constructing locally efficient estimators and conducting extensive simulations. We demonstrate the proposed methods in two real-world applications. *(abstract via OpenAlex)*
SMMR Survival Analysis Longitudinal Data Missing Data & Measurement Error · 发表于 2025-01-31
Tsung-I Lin, Wan-Lun Wang
摘要
The article proposes a robust approach to jointly modeling multiple repeated clinical measures with intricate features. More specifically, we aim to expand the scope of the multivariate linear mixed model by using the multivariate contaminated normal distribution. The proposed model, called the multivariate contaminated normal linear mixed model with censored and missing responses (MCNLMM-CM), is designed to handle minor outliers effectively, while simultaneously accommodating censored measurements and intermittent missing responses. An expectation conditional maximization either algorithm is developed to estimate the parameters of the proposed model in situations involving missing at random responses. We also provide techniques for approximating the asymptotic standard errors of the parameters, recovering censored data, imputing missing values, and identifying outliers. A simulation study is conducted to evaluate the finite-sample properties of the parameter estimators and demonstrate the superior performance of the proposed model compared to existing models. The proposed methodology is inspired by and applied to data from the Alzheimer’s disease neuroimaging initiative cohort study, which involves longitudinal clinical measurements of patients with mild cognitive impairment.
SMMR Missing Data & Measurement Error High-Dimensional · 发表于 2025-01-31
Lars Leonardus Joannes van der Burg, Hein Putter, Henning Baldauf, Jürgen Sauter, Johannes Schetelig, Liesbeth C de Wreede, Stefan Böhringer
摘要
Missing data problems are common in biological, high-dimensional data, where data can be partially or completely missing. Algorithms have been developed to reconstruct the missing values by means of imputation or expectation-maximization algorithms. For missing data problems, it has been suggested that the regression model of interest should be incorporated into the imputation procedure to reduce bias of the regression coefficients. We here consider a challenging missing data problem, where diplotypes of the KIR loci are to be reconstructed. These loci are difficult to genotype, resulting in ambiguous genotype calls. We extend a previously proposed expectation-maximization algorithm by incorporating a potentially high-dimensional regression model to model the outcome. Three strategies are evaluated: (1) only allelic predictors, (2) allelic predictors and forward-backward selection on haplotype predictors, and (3) penalized regression on a saturated model. In a simulation study, we compared these strategies with a baseline expectation-maximization algorithm without outcome model. For extreme choices of effect sizes and missingness levels, the outcome-based expectation-maximization algorithms outperformed the no-outcome expectation-maximization algorithm. However, in all other cases, the no-outcome expectation-maximization algorithm performed either superior or comparable to the three strategies, suggesting the outcome model can have a harmful effect. In a data analysis concerning death after allogeneic hematopoietic stem cell transplantation as a function of donor KIR genes, expectation-maximization algorithms with and without outcome showed very similar results. In conclusion, outcome based missing data models in the high-dimensional setting have to be used with care and are likely to lead to biased results.
JASA Causal Inference · 发表于 2025-01-31
Ting Ye, Kan Chen, Dylan Small
摘要
Does having firearms in the home increase suicide risk? To test this hypothesis, a matched case-control study can be performed, in which suicide case subjects are compared to living controls who are similar in observed covariates in terms of their retrospective exposure to firearms at home. In this application, cases can be defined using a broad case definition (suicide) or a narrow case definition (suicide occurred at home). The broad case definition offers a larger number of cases, but the narrow case definition may offer a larger effect size, which can reduce sensitivity to bias from unmeasured confounding. However, when the goal is to test whether there is a treatment effect based on the broad case definition, restricting to the narrow case definition may introduce selection bias (i.e., bias due to selecting samples based on characteristics affected by the treatment) because exposure to firearms in the home may affect the location of suicide and thus the type of a case a subject is. We propose a new sensitivity analysis framework for combining broad and narrow case definitions in matched case-control studies, that considers the unmeasured confounding bias and selection bias simultaneously. We develop a valid randomization-based testing procedure using only the narrow case matched sets when the effect of the unmeasured confounder on receiving treatment and the effect of the treatment on case definition among the always-cases are controlled by sensitivity parameters. We then use the Bonferroni method to combine the testing procedures using the broad and narrow case definitions. With the proposed methods, we find robust evidence that having firearms at home increases suicide risk. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*
Biometrics Genetics & Omics Testing & Uncertainty · 发表于 2025-01-31
Xiufan Yu, Linjun Zhang, Arun Srinivasan, Min-ge Xie, Lingzhou Xue
摘要
We introduce a novel meta-analysis framework to combine dependent tests under a general setting, and utilize it to synthesize various microbiome association tests that are calculated from the same dataset. Our development builds upon the classical meta-analysis methods of aggregating P-values and also a more recent general method of combining confidence distributions, but makes generalizations to handle dependent tests. The proposed framework ensures rigorous statistical guarantees, and we provide a comprehensive study and compare it with various existing dependent combination methods. Notably, we demonstrate that the widely used Cauchy combination method for dependent tests, referred to as the vanilla Cauchy combination in this article, can be viewed as a special case within our framework. Moreover, the proposed framework provides a way to address the problem when the distributional assumptions underlying the vanilla Cauchy combination are violated. Our numerical results demonstrate that ignoring the dependence among the to-be-combined components may lead to a severe size distortion phenomenon. Compared to the existing P-value combination methods, including the vanilla Cauchy combination method and other methods, the proposed combination framework is flexible and can be adapted to handle the dependence accurately and utilizes the information efficiently to construct tests with accurate size and enhanced power. The development is applied to the microbiome association studies, where we aggregate information from multiple existing tests using the same dataset. The combined tests harness the strengths of each individual test across a wide range of alternative spaces, enabling more efficient and meaningful discoveries of vital microbiome associations.
Biometrics Genetics & Omics Bayesian Methods · 发表于 2025-01-31
Yanghong Guo, Lei Yu, Lei Guo, Lin Xu, Qiwei Li
摘要
The abundance of various cell types can vary significantly among patients with varying phenotypes and even those with the same phenotype. Recent scientific advancements provide mounting evidence that other clinical variables, such as age, gender, and lifestyle habits, can also influence the abundance of certain cell types. However, current methods for integrating single-cell-level omics data with clinical variables are inadequate. In this study, we propose a regularized Bayesian Dirichlet-multinomial regression framework to investigate the relationship between single-cell RNA sequencing data and patient-level clinical data. Additionally, the model employs a novel hierarchical tree structure to identify such relationships at different cell-type levels. Our model successfully uncovers significant associations between specific cell types and clinical variables across three distinct diseases: pulmonary fibrosis, COVID-19, and non-small cell lung cancer. This integrative analysis provides biological insights and could potentially inform clinical interventions for various diseases.
Biometrics Clinical Trials Causal Inference Missing Data & Measurement Error · 发表于 2025-01-28
Camila Olarte Parra, Rhian M Daniel, David Wright, Jonathan W Bartlett
摘要
The ICH E9 addendum on estimands in clinical trials provides a framework for precisely defining the treatment effect that is to be estimated, but says little about estimation methods. Here, we report analyses of a clinical trial in type 2 diabetes, targeting the effects of randomized treatment, handling rescue treatment and discontinuation of randomized treatment using the so-called hypothetical strategy. We show how this can be estimated using mixed models for repeated measures, multiple imputation, inverse probability of treatment weighting, G-formula, and G-estimation. We describe their assumptions and practical details of their implementation using packages in R. We report the results of these analyses, broadly finding similar estimates and standard errors across the estimators. We discuss various considerations relevant when choosing an estimation approach, including computational time, how to handle missing data, whether to include post intercurrent event data in the analysis, whether and how to adjust for additional time-varying confounders, and whether and how to model different types of intercurrent event data separately.
SiM Diagnostics & Prediction Machine Learning · 发表于 2025-01-26
Alex Carriero, Kim Luijken, Anne de Hond, Karel G. M. Moons, Ben van Calster, Maarten van Smeden
摘要
Introduction: Risk prediction models are increasingly used in healthcare to aid in clinical decision‐making. In most clinical contexts, model calibration (i.e., assessing the reliability of risk estimates) is critical. Data available for model development are often not perfectly balanced with the modeled outcome (i.e., individuals with vs. without the event of interest are not equally prevalent in the data). It is common for researchers to correct for class imbalance, yet, the effect of such imbalance corrections on the calibration of machine learning models is largely unknown. Methods: We studied the effect of imbalance corrections on model calibration for a variety of machine learning algorithms. Using extensive Monte Carlo simulations we compared the out‐of‐sample predictive performance of models developed with an imbalance correction to those developed without a correction for class imbalance across different data‐generating scenarios (varying sample size, the number of predictors, and event fraction). Our findings were illustrated in a case study using MIMIC‐III data. Results: In all simulation scenarios, prediction models developed without a correction for class imbalance consistently had equal or better calibration performance than prediction models developed with a correction for class imbalance. The miscalibration introduced by correcting for class imbalance was characterized by an over‐estimation of risk and was not always able to be corrected with re‐calibration. Conclusion: Correcting for class imbalance is not always necessary and may even be harmful to clinical prediction models which aim to produce reliable risk estimates on an individual basis.
SiM Other · 发表于 2025-01-26
Jingmao Li, Qingzhao Zhang, Shuangge Ma, Kuangnan Fang, Yaqing Xu
摘要
In biomedical studies, gene‐environment (G‐E) interactions have been demonstrated to have important implications for analyzing disease outcomes beyond the main G and main E effects. Many approaches have been developed for G‐E interaction analysis, yielding important findings. However, hierarchical multi‐label classification, which provides insightful information on disease outcomes, remains unexplored in G‐E analysis literature. Moreover, unlabeled data are commonly observed in practical settings but omitted by many existing methods of hierarchical multi‐label classification. In this study, we consider a semi‐supervised scenario and develop a novel approach for the two‐layer hierarchical response with G‐E interactions. A two‐step penalized estimation is then proposed using an efficient expectation‐maximization (EM) algorithm. Simulation shows that it has superior performance in classification and feature selection. The analysis of The Cancer Genome Atlas (TCGA) data on lung cancer demonstrates the practical utility of the proposed method. Overall, this study can fill the important knowledge gap in G‐E interaction analysis by providing a widely applicable framework for hierarchical multi‐label classification of complex disease outcomes.
SiM Survival Analysis Longitudinal Data Bayesian Methods · 发表于 2025-01-26
Kelin Zhong, Fernanda L. Schumacher, Luis M. Castro, Víctor H. Lachos
摘要
The use of mixed‐effect models to understand the evolution of the human immunodeficiency virus (HIV) and the progression of acquired immune deficiency syndrome (AIDS) has been the cornerstone of longitudinal data analysis in recent years. However, data from HIV/AIDS clinical trials have several complexities. Some of the most common recurrences are related to the situation where the HIV viral load can be undetectable, and the measures of the patient can be registered irregularly due to some problems in the data collection. Although censored mixed‐effects models assuming conditionally independent normal random errors are commonly used to analyze this data type, this model may need to be more appropriate for accommodating outlying observations and responses recorded at irregular intervals. Consequently, in this paper, we propose a Bayesian analysis of censored linear mixed‐effects models that replace Gaussian assumptions with a flexible class of distributions, such as the scale mixture of normal family distributions, considering a damped exponential correlation structure that was employed to account for within‐subject autocorrelation among irregularly observed measures. For this complex structure, Stan's default No‐U‐Turn sampler is utilized to obtain posterior simulations. The feasibility of the proposed methods was demonstrated through several simulation studies and their application to two AIDS case studies.
SiM Longitudinal Data Survival Analysis Bayesian Methods · 发表于 2025-01-26
Danilo Alvares, Jessica K. Barrett, François Mercier, Spyros Roumpanis, Sean Yiu, Felipe Castro, Jochen Schulze, Yajing Zhu
摘要
Predicting cancer‐associated clinical events is challenging in oncology. In Multiple Myeloma (MM), a cancer of plasma cells, disease progression is determined by changes in biomarkers, such as serum concentration of the paraprotein secreted by plasma cells (M‐protein). Therefore, the time‐dependent behavior of M‐protein and the transition across lines of therapy (LoT), which may be a consequence of disease progression, should be accounted for in statistical models to predict relevant clinical outcomes. Furthermore, it is important to understand the contribution of the patterns of longitudinal biomarkers, upon each LoT initiation, to time‐to‐death or time‐to‐next‐LoT. Motivated by these challenges, we propose a Bayesian joint model for trajectories of multiple longitudinal biomarkers, such as M‐protein, and the competing risks of death and transition to the next LoT. Additionally, we explore two estimation approaches for our joint model: simultaneous estimation of all parameters (joint estimation) and sequential estimation of parameters using a corrected two‐stage strategy aiming to reduce computational time. Our proposed model and estimation methods are applied to a retrospective cohort study from a real‐world database of patients diagnosed with MM in the US from January 2015 to February 2022. We split the data into training and test sets in order to validate the joint model using both estimation approaches and make dynamic predictions of times until clinical events of interest, informed by longitudinally measured biomarkers and baseline variables available up to the time of prediction.
SiM Clinical Trials Bayesian Methods · 发表于 2025-01-25
Ping Gao
摘要
Sample size re‐estimation (SSR) is perhaps the most used adaptive procedure in both frequentist and Bayesian adaptive designs for clinical trials. The primary focus of all current frequentist and Bayesian SSR procedures is type I error control. We propose a hybrid frequentist‐Bayesian SSR approach that focuses on optimizing operating characteristics (OC), which uses simulations to investigate the associated OC and adjusts accordingly. The hybrid approach incorporates the Bayesian predictive power into the frequentist framework of SSR. Simulations show that the hybrid approach can substantially outperform popular frequentist type error‐focused SSR procedure. The hybrid approach can substantially improve the effectiveness of SSR using Bayesian predictive power.
SiM Bayesian Methods · 发表于 2025-01-25
Thanthirige Lakshika M. Ruberu, Danielle Braun, Giovanni Parmigiani, Swati Biswas
摘要
Multi‐gene panel testing allows efficient detection of pathogenic variants in cancer susceptibility genes including moderate‐risk genes such as ATM and PALB2. A growing number of studies examine the risk of breast cancer (BC) conferred by pathogenic variants of these genes. A meta‐analysis combining the reported risk estimates can provide an overall estimate of age‐specific risk of developing BC, that is, penetrance for a gene. However, estimates reported by case–control studies often suffer from ascertainment bias. Currently, there is no method available to adjust for such bias in this setting. We consider a Bayesian random effect meta‐analysis method that can synthesize different types of risk measures and extend it to incorporate studies with ascertainment bias. This is achieved by introducing a bias term in the model and assigning appropriate priors. We validate the method through a simulation study and apply it to estimate BC penetrance for carriers of pathogenic variants in the ATM and PALB2 genes. Our simulations show that the proposed method results in more accurate and precise penetrance estimates compared to when no adjustment is made for ascertainment bias or when such biased studies are discarded from the analysis. The overall estimated BC risk for individuals with pathogenic variants are (1) 5.77% (3.22%–9.67%) by age 50 and 26.13% (20.31%–32.94%) by age 80 for ATM; (2) 12.99% (6.48%–22.23%) by age 50, and 44.69% (34.40%–55.80%) by age 80 for PALB2. The proposed method allows meta‐analyses to include studies with ascertainment bias, resulting in inclusion of more studies and thereby more accurate estimates.
SiM Testing & Uncertainty · 发表于 2025-01-24
Óscar Lado‐Baleato, Javier Roca‐Pardiñas, Carmen Cadarso‐Suárez, Francisco Gude
摘要
Correlated clinical measurements are routinely interpreted via comparisons with univariate reference intervals examined side by side. Multivariate reference regions (MVRs), i.e., regions that characterize the distribution of multivariate results, have been proposed as a more adequate interpretation tool in such situations. However, MVR estimation methods have not yet been fully developed and are rarely used by physicians. The multivariate distribution of correlated measurements might change with certain patient characteristics (e.g., age or gender), and their effect on the shape of an MVR can be complex, involving interaction terms. For instance, the reference region shape for a given set of continuous covariates might vary across groups with respect to the value of a categorical variable. This paper examines the use of a bootstrap‐based hypothesis test for examining the effect of covariates on bivariate reference regions, testing the effect of factor‐by‐region interactions. An estimation algorithm based on smoothing splines was used to construct the bivariate reference region for a pediatric anthropometric dataset, and the bootstrapping procedure was used to determine the effect of age and gender on the shape of the reference region. (Height, weight) bivariate distribution was shown to depend on the interaction between age and gender. The bootstrapping procedure confirmed that a bivariate growth chart is desirable over univariate age‐gender body mass index (BMI) percentile curves. Whereas the well‐known BMI criterion detects only two atypical situations (i.e., underweight, overweight), the bootstrap‐tested bivariate reference region detected abnormally large or small body frames for different ages and genders.
SiM Functional Data Survival Analysis · 发表于 2025-01-24
Jiahui Feng, Haolun Shi, Da Ma, Mirza Faisal Beg, Jiguo Cao
摘要
Brain imaging data is one of the primary predictors for assessing the risk of Alzheimer's disease (AD). This study aims to extract image‐based features associated with the possibly right‐censored time‐to‐event outcomes and to improve predictive performance. While the functional proportional hazards model is well‐studied in the literature, these studies often do not consider the existence of patients who have a very low risk and are approximately insusceptible to AD. We introduce a functional mixture cure rate model that extends the proportional hazards model by allowing a proportion of event‐free patients. We propose a novel supervised functional principal component analysis (sFPCA) method to extract image features associated with AD risk while accounting for the complexity arising from right censoring. The proposed method accommodates the irregular boundary issue inherent in brain images with bivariate splines over triangulations. We demonstrate the advantages of the proposed method through extensive simulation studies and provide an application to the Alzheimer's Disease Neuroimaging Initiative (ADNI) study.
SiM Clinical Trials · 发表于 2025-01-24
Liwen Wu, Junyao Wang, Abdus S. Wahed
摘要
Clinical trials are often designed based on limited information about effect sizes and precision parameters with risks of underpowered studies. This is more problematic for SMARTs where strategy effects are based on sequences of treatments. Sample size adjustment offers flexibility through re‐estimating sample size during the trial to ensure adequate power at the final analysis. While this adaptation is common for standard clinical trials, corresponding methods to perform sample size adjustment have not been adapted to SMARTs. In this paper, we propose a sample size adjustment procedure for SMARTs. Sample sizes are re‐calculated at the interim analysis based on the conditional power derived from a bivariate non‐central chi‐square distribution. We demonstrate through simulation studies that even with an underpowered initial sample size due to miss‐specified parameters at the design stage, the proposed method can maintain desirable power at the end of the study, and additional resources are only invested in trials that show promising conditional power at the interim analysis.
SiM Missing Data & Measurement Error Longitudinal Data Clinical Trials · 发表于 2025-01-24
Suzie Cro, Matteo Quartagno, Ian R. White, James R. Carpenter
摘要
Introduction: In clinical trials, a treatment policy strategy is often used to handle treatment nonadherence. However, estimation in this context is complicated when data are missing after treatment deviation. Reference‐based multiple imputation has been developed for the analysis of a longitudinal continuous outcome in this setting. It has been shown that Rubin's variance estimator ensures that the proportional loss of information due to missing data is approximately the same as that seen in analysis under the missing‐at‐random assumption for a broad range of commonly used reference‐based alternatives; that is it is information anchored. However, the best way to implement reference‐based multiple imputation for longitudinal binary data is unclear. Methods: We formulate and describe two algorithms for implementing reference‐based multiple imputation for longitudinal binary outcome data using: (i) joint modeling with the multivariate normal distribution and an adaptive rounding algorithm and (ii) joint modeling with a latent multivariate normal model. A simulation study was performed to compare the properties of the two methods. Results: Across the broad range of scenarios evaluated, the latent normal approach typically gave slightly less bias; both methods provided approximately information anchored inference. The advantage of the latent normal approach was more marked with a rarer outcome. However, both approaches may not perform satisfactorily if the outcome prevalence is very rare, that is, . Discussion: Reference‐based multiple imputation provides a practical information anchored tool for inferences about the treatment effect for a treatment policy estimand with a longitudinal binary outcome. The latent multivariate normal model is the preferred implementation.
SiM Clinical Trials Bayesian Methods High-Dimensional · 发表于 2025-01-24
Adrien Ollier, Pavel Mozgunov
摘要
There is a growing number of Phase I dose‐finding studies that use a model‐based approach, such as the CRM or the EWOC method to estimate the dose‐toxicity relationship. It is common to assume that all patients will have similar toxicity risk given the dose regardless of patients' individual characteristics. In many trials, however, some patients' covariates (e.g., a concomitant drug assigned by a clinician) might have an impact on the dose‐toxicity relationship. In this work, motivated by a real trial, we evaluate an impact of taking into account (or omitting) some patients' covariates on the individual target dose recommendations and patients' safety in Phase I model‐based dose‐finding study. We investigate several variable penalisation criteria and found that, for continuous and binary covariates, omitting a prognostic covariate leads to a drastically low proportion of correct selections and an increase of overdosing. At the same time, including a covariate can lead to good operating characteristics in all scenarios but can sometimes slightly decrease the proportion of good selections and increase the overdosing. To tackle this, we propose to use a Bayesian Lasso Bayesian Logistic Regression Model (BLRM) and Spike‐and‐Slab BLRM. We have found that the BLRM coupled to the Bayesian LASSO and the BLRM with Spike‐and‐Slab are on average better appropriate to consider variable inclusion.
SiM Survival Analysis · 发表于 2025-01-24
Ruiqian Wu, Ying Zhang, Giorgos Bakoyannis
摘要
The semi‐competing risks data model is a special type of disease‐state model that focuses on studying the association between an intermediate event and a terminal event and proves to be a useful tool in modeling disease progression. The study of the semi‐competing risk data model not only allows us to evaluate whether a disease episode is related to death but also provides a toolkit to predict death, given that the episode occurred at a certain time. However, the computation of the semi‐competing risk models is a numerically challenging task. The Gamma‐Frailty conditional Markov model has been shown to be an efficient computation model for studying semi‐competing risks data. Building on recent advances in studying semi‐competing risks data, this work proposes a non‐parametric pseudo‐likelihood method equipped with an EM‐like algorithm to study semi‐competing risks data with event misascertainment under the restricted Gamma‐Frailty conditional Markov model. A thorough simulation study is conducted to demonstrate the inference validity of the proposed method and its numerical stability. The proposed method is applied to a large HIV cohort study, EA‐IeDEA, that has a severe death under‐reporting issue to assess the degree of adverse impact of the interruption of ART care on HIV mortality.
SiM Causal Inference Survival Analysis Nonparametric & Semiparametric · 发表于 2025-01-24
Yen‐Tsung Huang, Ju‐Sheng Hong
摘要
A causal mediation model with multiple time‐to‐event mediators is exemplified by the natural course of human disease marked by sequential milestones with a time‐to‐event nature. For example, from hepatitis B infection to death, patients may experience intermediate events such as liver cirrhosis and liver cancer. The sequential events of hepatitis, cirrhosis, cancer, and death are susceptible to right censoring; moreover, the latter events may preclude the former events. Casting the natural course of human diseases in the framework of causal mediation modeling, we establish a model with intermediate and terminal events as the mediators and outcomes, respectively. We define the interventional analog of path‐specific effects (iPSEs) as the effect of an exposure on a terminal event mediated (or not mediated) by any combination of intermediate events without parametric models. The expression of a counting process‐based counterfactual hazard is derived under the sequential ignorability assumption. We employ composite nonparametric likelihood estimation to obtain maximum likelihood estimators for the counterfactual hazard and iPSEs. Our proposed estimators achieve asymptotic unbiasedness, uniform consistency, and weak convergence. Applying the proposed method, we show that hepatitis B induced mortality is mostly mediated through liver cancer and/or cirrhosis whereas hepatitis C induced mortality may be through extrahepatic diseases.
SiM Causal Inference Survival Analysis · 发表于 2025-01-24
Na Bo, Jong‐Hyeon Jeong, Erick Forno, Ying Ding
摘要
An important aspect of precision medicine focuses on characterizing diverse responses to treatment due to unique patient characteristics, also known as heterogeneous treatment effects (HTE) or individualized treatment effects (ITE), and identifying beneficial subgroups with enhanced treatment effects. Estimating HTE with right‐censored data in observational studies remains challenging. In this paper, we propose a pseudo‐ITE‐based framework for analyzing HTE in survival data, which includes a group of meta‐learners for estimating HTE, a variable importance metric for identifying predictive variables to HTE, and a data‐adaptive procedure to select subgroups with enhanced treatment effects. We evaluate the finite sample performance of the framework under various observational study settings. Furthermore, we applied the proposed methods to analyze the treatment heterogeneity of a written asthma action plan (WAAP) on time‐to‐ED (Emergency Department) return due to asthma exacerbation using a large asthma electronic health records dataset with visit records expanded from pre‐ to post‐COVID‐19 pandemic. We identified vulnerable subgroups of patients with poorer asthma outcomes but enhanced benefits from WAAP and characterized patient profiles. Our research provides valuable insights for healthcare providers on the strategic distribution of WAAP, particularly during disruptive public health crises, ultimately improving the management and control of pediatric asthma.
SiM Other · 发表于 2025-01-24
Ao Sun, Xiao‐Hua Zhou
摘要
The ideal evaluation of diagnostic test performance requires a reference test that is free of errors. However, for many diseases, obtaining such a “gold standard” reference is either impossible or prohibitively expensive. Estimating test accuracy in the absence of a gold standard is therefore a significant challenge. In this article, we introduce and categorize existing methods for evaluating diagnostic tests without a gold standard, considering factors such as the type and number of tests, as well as the structure of the observed data. For each method, we provide a comprehensive introduction and analysis of its underlying assumptions, model architecture, identifiability, estimation techniques, and inference procedures. We use R to conduct simulations for widely applicable models, validating assumptions, comparing models, and assessing their reliability. Additionally, we present real‐world examples along with the corresponding R code for these models, enabling readers to better understand how to apply them effectively. Beyond diagnostic medicine, we underscore that the issue of imperfect gold standards affects other fields, drawing parallels to the noisy label problem in machine learning. By highlighting similarities and differences across these domains, we open pathways for further research. The primary aim of this article is to consolidate existing methods for assessing test accuracy in the absence of a gold standard and to provide practical guidance for researchers seeking to apply these methods effectively.
Biometrics Meta-Analysis & Data Integration Bayesian Methods · 发表于 2025-01-24
Jennifer F Bobb, Stephen J Mooney, Maricela Cruz, Anne Vernez Moudon, Adam Drewnowski, David Arterburn, Andrea J Cook
摘要
Distributed lag models (DLMs) estimate the health effects of exposure over multiple time lags prior to the outcome and are widely used in time series studies. Applying DLMs to retrospective cohort studies is challenging due to inconsistent lengths of exposure history across participants, which is common when using electronic health record databases. A standard approach is to define subcohorts of individuals with some minimum exposure history, but this limits power and may amplify selection bias. We propose alternative full-cohort methods that use all available data while simultaneously enabling examination of the longest time lag estimable in the cohort. Through simulation studies, we find that restricting to a subcohort can lead to biased estimates of exposure effects due to confounding by correlated exposures at more distant lags. By contrast, full-cohort methods that incorporate multiple imputation of complete exposure histories can avoid this bias to efficiently estimate lagged and cumulative effects. Applying full-cohort DLMs to a study examining the association between residential density (a proxy for walkability) over 12 years and body weight, we find evidence of an immediate effect in the prior 1-2 years. We also observed an association at the maximal lag considered (12 years prior), which we posit reflects an earlier ($\ge$12 years) or incrementally increasing prior effect over time. DLMs can be efficiently incorporated within retrospective cohort studies to identify critical windows of exposure.
SiM Nonparametric & Semiparametric · 发表于 2025-01-24
Shengxin Tu, Chun Li, Bryan E. Shepherd
摘要
Clustered data are common in practice. Clustering arises when subjects are measured repeatedly, or subjects are nested in groups (e.g., households, schools). It is often of interest to evaluate the correlation between two variables with clustered data. There are three commonly used Pearson correlation coefficients (total, between‐, and within‐cluster), which together provide an enriched perspective of the correlation. However, these Pearson correlation coefficients are sensitive to extreme values and skewed distributions. They also vary with data transformation, which is arbitrary and often difficult to choose, and they are not applicable to ordered categorical data. Current nonparametric correlation measures for clustered data are only for the total correlation. Here we define population parameters for the between‐ and within‐cluster Spearman rank correlations. The definitions are natural extensions of the Pearson between‐ and within‐cluster correlations to the rank scale. We show that the total Spearman rank correlation approximates a linear combination of the between‐ and within‐cluster Spearman rank correlations, where the weights are functions of rank intraclass correlations of the two random variables. We also discuss the equivalence between the within‐cluster Spearman rank correlation and the covariate‐adjusted partial Spearman rank correlation. Furthermore, we describe estimation and inference for the three Spearman rank correlations, conduct simulations to evaluate the performance of our estimators, and illustrate their use with data from a longitudinal biomarker study and a clustered randomized trial.
SiM Survival Analysis Spatial & Time Series Bayesian Methods · 发表于 2025-01-24
Yuliya Leontyeva, Yuxin Huang, Susanna Cramb, Jessica Cameron, Peter Baade, Kerrie Mengersen, Helen Thompson
摘要
To date, there have not been any population‐based cancer studies quantifying geographical patterns of the loss in life expectancy (LLE) and crude probability of death due to cancer (). These absolute measures of survival are complementary to the more typically used relative measures of excess mortality and relative survival, and, together, they provide a fuller understanding of geographical disparities in survival outcomes for cancer patients. We propose using a spatially flexible parametric relative survival model in the Bayesian framework, which allows for the inclusion of spatial effects in hazard‐level model components. The relative survival framework is the preferred approach to analyze cancer survival data because it does not require information on the cause of death, and the Bayesian spatial modeling approach allows complex and robust small‐area estimation. The calculation of spatial estimates for LLE and are demonstrated using publicly available simulated datasets. The associated computer program scripts are available to support the understanding and implementation of our methodology in other spatial cancer modelling applications.
SiM Bayesian Methods · 发表于 2025-01-24
Zifei Han, Qiang Zhang, Ram Tiwari, Tianyu Bai
摘要
The recent U.S. Food and Drug Administration guidance on complex innovative trial designs acknowledges the use of Bayesian strategies to incorporate historical information based on clinical expertise and data similarity. Also, data from multiple previous studies with similar settings often qualify for historical borrowing. Although several classes of informative priors can semi‐automatically leverage historical information based on data compatibility, it is common that some exogenous factors, such as the year of patient enrollment, can also influence the relevance of each historical study to the current trial. Consequently, a natural a priori ordering among historical trials often arises, a constraint that many current informative priors fail to accommodate. Motivated by a pediatric lupus clinical study and an oncology trial, we introduce a variant of the power prior, named the ordered normalized power prior, which ensures a targeted order restriction on the power parameters and maintains data‐adaptive borrowing. We further explore and compare two distinct normalization strategies and outline computational details with efficient sampling algorithms. The clinical datasets mentioned are analyzed, and extensive simulations are conducted for comparison. An efficient implementation is provided in our updated package NPP available on the Comprehensive R Archive Network.
SiM Survival Analysis Nonparametric & Semiparametric · 发表于 2025-01-24
Zhao Da, Xia Xiaochao, Li Jialiang
摘要
The additive hazard model, which focuses on risk differences rather than risk ratios, has been widely applied in practice. In this paper, we consider an additive hazard model with varying coefficients to analyze recurrent events data. The model allows for both varying and constant coefficients. We first propose an estimating equation‐based approach with spline basis smoothing for all functional coefficients. Then, we provide theoretical justifications for the resulting estimates, including consistency, rate of convergence, and asymptotic distribution. Furthermore, we construct a Cramér–von Mises test procedure to investigate whether the functional coefficients should be treated as constant, and its asymptotic null distribution is also derived. Extensive simulation experiments are conducted to evaluate the finite‐sample performance of the proposed approaches. A Chronic Granulotamous Disease data set was analyzed to illustrate our methodology.
SiM Causal Inference · 发表于 2025-01-24
Chengyao Tang, Yi Zhou, Ao Huang, Satoshi Hattori
摘要
In estimating the average treatment effect in observational studies, the influence of confounders should be appropriately addressed. To this end, the propensity score is widely used. If the propensity scores are known for all the subjects, bias due to confounders can be adjusted by using the inverse probability weighting (IPW) by the propensity score. Since the propensity score is unknown in general, it is usually estimated by the parametric logistic regression model with unknown parameters estimated by solving the score equation under the strongly ignorable treatment assignment (SITA) assumption. Violation of the SITA assumption and/or misspecification of the propensity score model can cause serious bias in estimating the average treatment effect (ATE). To relax the SITA assumption, the IPW estimator based on the outcome‐dependent propensity score has been successfully introduced. However, it still depends on the correctly specified parametric model and its identification. In this paper, we propose a simple sensitivity analysis method for unmeasured confounders. In the standard practice, the estimating equation is used to estimate the unknown parameters in the parametric propensity score model. Our idea is to make inferences on the (ATE) by removing restrictive parametric model assumptions while still utilizing the estimating equation. Using estimating equations as constraints, which the true propensity scores asymptotically satisfy, we construct the worst‐case bounds for the ATE with linear programming. Differently from the existing sensitivity analysis methods, we construct the worst‐case bounds with minimal assumptions. We illustrate our proposal by simulation studies and a real‐world example.
SiM Bayesian Methods Genetics & Omics Networks & Graphs · 发表于 2025-01-24
Kevin C. Lutz, Michael L. Neugent, Tejasv Bedi, Nicole J. De Nisco, Qiwei Li
摘要
Advances in next‐generation sequencing technology have enabled the high‐throughput profiling of metagenomes and accelerated microbiome studies. Recently, there has been a rise in quantitative studies that aim to decipher the microbiome co‐occurrence network and its underlying community structure based on metagenomic sequence data. Uncovering the complex microbiome community structure is essential to understanding the role of the microbiome in disease progression and susceptibility. Taxonomic abundance data generated from metagenomic sequencing technologies are high‐dimensional and compositional, suffering from uneven sampling depth, over‐dispersion, and zero‐inflation. These characteristics often challenge the reliability of the current methods for microbiome community detection. To study the microbiome co‐occurrence network and perform community detection, we propose a generalized Bayesian stochastic block model that is tailored for microbiome data analysis where the data are transformed using the recently developed modified centered‐log ratio transformation. Our model also allows us to leverage taxonomic tree information using a Markov random field prior. The model parameters are jointly inferred by using Markov chain Monte Carlo sampling techniques. Our simulation study showed that the proposed approach performs better than competing methods even when taxonomic tree information is non‐informative. We applied our approach to a real urinary microbiome dataset from postmenopausal women. To the best of our knowledge, this is the first time the urinary microbiome co‐occurrence network structure in postmenopausal women has been studied. In summary, this statistical methodology provides a new tool for facilitating advanced microbiome studies.