研究园简易版 · 经典书籍 · 期刊文章 · 有原文 · 完整版(可多选筛选、搜索) · 首页
2026 年的期刊文章 · 第 14 页
共 1756 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2026-07-24 到 2026-07-15。
按期刊、按主题看 →
JSSAM Causal Inference Missing Data & Measurement Error Networks & Graphs · 发表于 2026-07-24
Not available
摘要
Missing data poses fundamental challenges for causal discovery in observational studies, particularly when inferring directed acyclic graphs that represent underlying cause-and-effect relationships. Traditional approaches to handling missing data, such as complete-case analysis, can introduce biases and reduce the reliability of causal inferences. While multiple imputation via chained equations represents the current standard for handling missing data, its effectiveness for causal discovery is limited by both the absence of complete observations and the lack of principled methods for aggregating causal graphs across imputed datasets. This work introduces a novel unified framework that integrates causal structure learning directly into the imputation process. The key innovation lies in using preliminary adjacency information to estimate missingness propensity scores, which are then incorporated into the imputation model together with source indicator variables to generate more accurate estimates of missing values. The conditional independence test is implemented with source indicators included in the conditioning set, and by pooling test statistics rather than graph structures across multiple imputations, the proposed approach addresses the traditional challenge of graph aggregation while maintaining statistical rigor. Simulation studies demonstrate that the proposed framework achieves better performance compared to the traditional approach across all relevant metrics. The proposed framework is adapted to handle complex survey data by incorporating survey weights and survey design in both data imputation and conditional independence tests for causal discovery. When applied to childhood anthropometric survey data from Mali, the proposed approach produces results that better align with expected outcomes. By combining imputation and causal discovery in a unified process while accounting for survey design, it provides survey researchers with a practical tool for integrating and analyzing incomplete, complex survey data from multiple sources or time points.
JRSSA Clinical Trials High-Dimensional · 发表于 2026-07-24
Not available
Biometrics Bayesian Methods · 发表于 2026-07-24
Siquan Wang, Qixuan Chen, Matthew S Perzanowski, Adnan Divjan, Andrew Gelman
摘要
Serial dilution assays are essential tools in biomedical research. Current analytical methods work well when measurements are in the steep part of the calibration curve but are not so efficient when measurements are near the upper and lower bounds. Moreover, sample contamination is common in serial dilution assays and can result in problematic concentration estimation. To address these challenges, we propose a Bayesian contamination model that simultaneously flags contaminated samples and estimates the concentrations of unknown samples, while quantifying estimation uncertainty. In extensive simulation studies under various contamination scenarios, our method consistently outperforms commonly used approaches. We further demonstrate a step-by-step application of a Bayesian workflow for modeling contamination using raw immunoassay data from the New York City Neighborhood Asthma and Allergy Study. This work advances previous research by introducing a practical method for detecting contaminated samples and estimating unknown concentrations in immunoassays and by presenting a general framework for incorporating unknown contamination processes into the Bayesian modeling workflow.
SiM Bayesian Methods Nonparametric & Semiparametric · 发表于 2026-07-23,收录于 2026-07-31
Huiqiong Li, Lu Luo, Min Wang, Niansheng Tang
摘要
High‐dimensional data arising in genomics, econometrics, and clinical medicine often exhibit substantial heterogeneity across multiple sources. While existing methods address multi‐source heterogeneity, they do not adequately accommodate the combined challenges of high dimensionality and between‐source heterogeneity. To address this gap, we propose a scalable Bayesian framework for multi‐source heterogeneous quantile regression with spike‐and‐slab priors for simultaneous parameter estimation and feature selection. To overcome computational challenges, we combine mean‐field variational inference with Laplace approximation and introduce a novel low‐rank variational correction strategy that substantially improves approximation accuracy and adaptability in high‐dimensional heterogeneous settings. This low‐rank correction effectively captures the underlying dependence structure, leading to more robust and efficient inference. For model assessment and diagnostic analysis, we further develop a Bayesian score test coupled with local influence analysis. Extensive simulation studies and an analysis of The Cancer Genome Atlas (TCGA) data from four cancer cohorts (ESCA, PAAD, PCPG, and READ) demonstrate the computational efficiency, scalability, and practical utility of the proposed method in high‐dimensional heterogeneous applications. The proposed low‐rank variational correction algorithms are implemented in the R package LRQVB , which is publicly available on CRAN.
SMMR Causal Inference · 发表于 2026-07-23
Mahsa Ashouri, Nicholas C Henderson
摘要
Developing tools for estimating heterogeneous treatment effects (HTEs) and individualized treatment effects has been an area of active research in recent years. While these tools have proven to be useful in many contexts, a concern when deploying such methods is the degree to which incorporating HTE into a prediction model provides an advantage over predictive methods that do not allow for heterogeneity in treatment effect across individuals. To address this concern, we propose a procedure that evaluates the extent to which an HTE model provides a predictive advantage. Specifically, our procedure targets the gain in predictive performance from using a flexible predictive model incorporating HTE versus an alternative partial linear model which is similar to the HTE-utilizing model except that it is constrained to not allow heterogeneity in treatment effect. By drawing upon recent work in using nested cross-validation techniques for prediction error inference, we generate confidence intervals for this measure of gain in predictive performance, which allows one to directly calculate the level at which one is confident of a substantial HTE-modeling gain in prediction—a quantity that we refer to as the predictive HTE p -value. Our procedure is generic and can be directly used to assess the benefit of modeling HTE for any method that incorporates treatment effect heterogeneity.
SiM Longitudinal Data · 发表于 2026-07-21,收录于 2026-08-07
George Stefan, Eleanor Pullenayegum
摘要
Longitudinal data are commonly encountered in biomedical research, including randomized trials and retrospective cohort studies. Subjects are typically followed over a period of time and may be scheduled for follow‐up at predetermined time points. However, subjects may miss their appointments or return at non‐specified times, leading to irregularity in the visit process. IIW‐GEEs have been developed as one method to account for this irregularity, whereby estimates from a visit intensity model are used as weights in a GEE model with an independent correlation structure. We show that currently available methods can be biased for situations in which the health outcome of interest may influence a subject's dropout from the study. We have extended the IIW‐GEE framework to adjust for informative dropout and have demonstrated via simulation studies that this bias can be significantly reduced. We have illustrated this method using the STAR*D clinical trial data, and observed that the disease trajectory was generally overestimated when informative dropout was not accounted for.
SiM Clinical Trials · 发表于 2026-07-21,收录于 2026-08-07
Math J. J. M. Candel, Gerard J. P. van Breukelen
摘要
For two‐treatment multicenter trials with a continuous outcome, designs are presented that minimize the number of subjects required or the research budget, when aiming for a desired power level. These designs optimize both the allocation ratio of participants between treatment and control condition within each center, and the balance between the number of centers and the number of participants per center. Because optimal designs rely on prior knowledge of variance parameters from the analysis model, which are unknown during the design stage, maximin designs are introduced. These designs guarantee a pre‐specified power level across plausible ranges of the unknown parameters and maximize power under worst‐case parameter values. The present study extends the existing literature on multicenter designs by deriving maximin designs not only for the case where the number of centers and the sample size per center can be optimized, but also when either the number of centers or the sample size per center is fixed due to practical constraints. An empirical example illustrates sample size calculation for such designs and potential budget savings. Additionally, an interactive R Shiny app has been developed to simplify sample size calculations for maximin multicenter trials.
Biometrika Testing & Uncertainty · 发表于 2026-07-21,收录于 2026-08-20
Congbin Xu, Yue Yu, Zhaojun Wang, Changliang Zou, Haojie Ren
摘要
(verbatim from the journal page)*: "Conformal prediction offers a distribution-free framework for constructing prediction sets with finite-sample coverage. Yet, efficiently leveraging multiple nonconformity scores to reduce set sizes remains an open challenge. Instead of selecting a single best score, this work introduces a principled aggregation strategy that intersects multiple conformal prediction sets whose confidence levels are optimally allocated to minimize empirical set size while maintaining asymptotic coverage. Two variants are developed to guarantee finite-sample coverage via sample splitting and full conformalization, respectively. An individualized allocation strategy is further proposed to promote local size efficiency while achieving asymptotic conditional coverage. Extensive experiments on synthetic and real-world datasets demonstrate that our approach achieves considerably smaller prediction sets than state-of-the-art baselines while maintaining valid coverage." *(Online: 21 July 2026)*
Biometrics Testing & Uncertainty High-Dimensional Genetics & Omics · 发表于 2026-07-20,收录于 2026-10-17
Charlie K. Guan and others
摘要
We propose one-at-a-time knockoffs (OATK), a new methodology for detecting important explanatory variables in linear regression problems (including ridge and lasso) while controlling the false discovery rate (FDR). For each explanatory variable, OATK generates a knockoff design matrix that preserves the Gram matrix by replacing one-at-a-time only the single corresponding column of the original design matrix. OATK is a substantial relaxation and simplification of the knockoff filter, which simultaneously generates all columns of the knockoff design matrix to satisfy a much larger set of constraints. To test each variable’s importance, statistics are then constructed by comparing the original versus knockoff coefficients. Under a mild correlation assumption on the original design matrix, we prove that OATK asymptotically controls the FDR at any desired level. Moreover, numerical results across a variety of simulation examples and a real genetics data set demonstrate that OATK provides good FDR control and regularly achieves (often substantially) higher power than existing approaches. In particular, in a genome-wide association study, OATK yielded more uniquely discovered genetic mutations associated with virological response to HIV therapeutics that were corroborated in clinical studies. Generating knockoffs one-at-a-time also has substantial computational advantages and facilitates additional enhancements, such as conditional calibration or derandomization, to further improve power and consistency of FDR control.
SiM Causal Inference Nonparametric & Semiparametric · 发表于 2026-07-20,收录于 2026-08-07
Li‐Sheng Zhuang, Jih‐Chang Yu, Yen‐Tsung Huang
摘要
The illness‐death model is widely used to characterize disease progression over time. Previous work focuses either on estimation without causal interpretation or on causal interpretation under a strong Markov assumption where the terminal event depends on the status but not the timing of the intermediate event. To bridge the research gap, we propose a new definition of counterfactual hazard that relaxes the Markov assumption by considering the entire history of the intermediate event. We derive an identification formula that involves an integral with respect to the probability density function of the intermediate event time. Direct and indirect effects refer to the influence of an exposure on the terminal event not mediated by, and mediated through, the intermediate event, respectively. We propose non‐parametric kernel estimators for the two effects and study their asymptotic properties. We conduct numerical simulations to examine the proposed estimators' finite‐sample performance. Applying the method to a hepatitis study where the Markov assumption is violated, we show that the effect of hepatitis C on mortality is not mediated through septicemia during the first 15 years of follow‐up.
JASA Clinical Trials · 发表于 2026-07-20
Marc Buyse
摘要
This book provides an informative and well-documented overview of recent advances in methods for clinical trials. The structure of the book is straightforward and makes for easy navigation across i... *(abstract via OpenAlex)*
Biometrics Latent Variables & Mixtures Bayesian Methods Nonparametric & Semiparametric · 发表于 2026-07-17
Shijie Yuan, Kevin Roberts, Noirrit Kiran Chandra, Yuan Ji, Peter Müller
摘要
We introduce a Bayesian nonparametric inference approach for aggregate adverse event (AE) monitoring across studies. The proposed model seamlessly integrates external data from historical trials to define a relevant background rate and accommodates varying levels of covariate granularity (ranging from patient-level details to study-level aggregated summary data). Inference is based on a covariate-dependent product partition model (PPMx). A central element of the model is the ability to group experimental units with similar profiles. We introduce a pairwise similarity measure, with which we set up a random partition of experimental units with comparable covariate profiles, thereby improving the precision of AE rate estimation. Importantly, the proposed framework supports real-time safety monitoring under blinding with a seamless transition to unblinded analyses when indicated. Using one case study and simulation studies, we demonstrate the model’s ability to detect safety signals and assess risk under diverse trial scenarios.
Biometrics Causal Inference Latent Variables & Mixtures · 发表于 2026-07-17,收录于 2026-10-02
Charlotte R Fowler, Xiaoxuan Cai, Habiballah Rahimi-Eichi, Lisa Dixon, Justin T Baker, Jukka-Pekka Onnela, Linda Valeri
摘要
*(publisher's abstract as deposited in Crossref, verbatim)* Individuals with bipolar disorder tend to cycle through disease states such as depression and mania. The heterogeneous nature of disease across states complicates the evaluation of interventions for bipolar disorder patients, as varied interventional success is observed within and across individuals. We hypothesize that disease state acts as an effect modifier for the causal effect of a given intervention on health outcomes. To address this dilemma, we propose an N-of-1 approach using an adapted auto-regressive hidden Markov model, applied to longitudinal mobile health data collected from individuals with bipolar disorder. This method allows us to identify a latent variable from mobile health data to be treated as an effect modifier between the exposure and outcome of interest while allowing for missing data. A counterfactual approach is employed for causal inference and to obtain a g-formula estimator to recover said effect. The performance of the proposed method is compared with naive approaches across extensive simulations and application to a multi-year smartphone study of bipolar patients, evaluating the individual effect of social activity on sleep duration across different latent disease states.
Biometrics Spatial & Time Series Latent Variables & Mixtures Bayesian Methods · 发表于 2026-07-17,收录于 2026-09-03
Srijato Bhattacharyya, Huiyan Sang, Bani Mallick
摘要
Spatial clustering is crucial in disease mapping by identifying subregions with different patterns of disease incidence or mortality. This study proposes a novel Bayesian spatial clustering method for multivariate spatial disease data, which allows for understanding geographic variations of multivariate disease patterns while accounting for both spatial information and dependence among multiple disease measurements. We develop a new random tele-connected graph partition model with an unknown number of clusters, which is capable of encouraging locally contiguous clusters and allowing for remote subregions to be clustered together. We use this prior in a Bayesian hierarchical model to detect spatial clusters and estimate cluster-specific disease patterns and dependence across the multivariate disease variables. We develop a tailored Markov chain Monte Carlo (MCMC) algorithm for posterior inference, utilizing efficient doubly split-merge samplers taking advantage of graph algorithms. We illustrate our method with simulation studies and apply it to investigate the clustering patterns of county-level prostate cancer mortality rate decline across six southern U.S. states from 1985 to 2014." (Online date: 2026-07-17.)
SMMR Nonparametric & Semiparametric Causal Inference Survival Analysis · 发表于 2026-07-17
Chathura Siriwardhana, Bakeerathan Gunaratnam, Karunarathna Bandara Kulasekera
摘要
Personalized treatment selection is often based on a single primary endpoint, even though complex diseases are typically monitored using multiple, heterogeneous outcomes. We develop a rank-based framework for individualized treatment selection that accommodates mixed collections of continuous, ordinal, binary, and right-censored survival outcomes. For each outcome and treatment, we summarize the conditional outcome distribution given patient covariates using clinically meaningful parameters, including conditional quantiles, survival probabilities, and tail probabilities at ordered thresholds, and assemble these summaries across treatment arms into vectors with a common structure. Each summary component induces a ranking of the treatments, and discrepancies between outcome-specific rankings and candidate ranking vectors are quantified using a Mahalanobis-type distance that incorporates user-specified outcome weights and an empirically estimated scaling matrix. The optimal treatment is defined as the arm with the most favorable overall rank. Outcome summaries are estimated using flexible tree-based ensemble methods, although other regression or machine-learning tools can be used without altering the decision rule. Monte Carlo simulations across a range of scenarios show reasonably high probabilities of correct treatment selection, with performance improving as the signal-to-noise ratio increases and only mildly affected by the choice of outcome weights. We illustrate the method using data from an AIDS clinical trial, deriving individualized antiretroviral regimen recommendations that balance CD4 response, CD4/CD8 ratio, and a composite time-to-event endpoint.
Biometrics Longitudinal Data Bayesian Methods Meta-Analysis & Data Integration · 发表于 2026-07-17,收录于 2026-09-03
Jialing Liu, Souradipto Ghosh Dastidar, Steffen Ventz
摘要
*(publisher's abstract as deposited in Crossref, verbatim)* The analysis of contemporary longitudinal data problems often involves high-dimensional measurements of time-course data collected on a small number of observations. The estimation of such models with limited sample sizes can pose substantial challenges and lead to highly variable estimates, unstable predictions, and low power. In such instances, it is natural to borrow information from additional datasets with similar, although not necessarily identical, covariate–outcome relations to improve inference in the target population. We develop a novel Bayesian transfer learning model for longitudinal data (BTLL). BTLL leverages mixture models for the discrepancies between the pivotal parameters of the outcome models of the target and source studies to enhance estimation accuracy and enable data-adaptive information borrowing. BTLL aims to minimize the negative transfer of information from source studies that would otherwise introduce large bias into inference in the target population. Through extensive simulations and real data applications, we show that BTLL improves the precision of parameter estimates in the target study substantially compared to alternative methods. Moreover, our BTLL reduces the estimation bias in heterogeneous settings when the outcome distributions of the target and source datasets deviate from each other. Motivated by the progressive nature of dementia and the importance of early detection, we develop a BTLL prediction model using datasets for individuals with mild cognitive impairment. We show that BTLL can leverage non-invasive biomarkers to identify subjects at high risk for dementia and consistently achieves the lowest prediction error.
SiM Latent Variables & Mixtures Genetics & Omics · 发表于 2026-07-16,收录于 2026-07-24
Not available
摘要
Electronic health records (EHRs) linked with familial relationship data offer a unique opportunity to investigate the genetic architecture of complex phenotypes at scale. However, existing heritability and coheritability estimation methods often fail to account for the intricacies of familial correlation structures, heterogeneity across phenotype types, and computational scalability. We propose a robust and flexible statistical framework for jointly estimating heritability and genetic correlation among continuous and binary phenotypes in EHR‐based family studies. Our approach builds on multi‐level latent variable models to decompose phenotypic covariance into interpretable genetic and environmental components, incorporating both within‐ and between‐family variations. We derive iteration algorithms based on generalized equation estimations (GEE) for estimation. Simulation studies under various parameter configurations demonstrate that our estimators are consistent and yield valid inference across a range of realistic settings. Applying our methods to real‐world EHR data from a large, urban health system, we identify significant genetic correlations between mental health conditions and endocrine/metabolic phenotypes, supporting hypotheses of shared etiology. This work provides a scalable and rigorous framework for coheritability analysis in high‐dimensional EHR data and facilitates the identification of shared genetic influences in complex disease networks.
SMMR Survival Analysis Nonparametric & Semiparametric · 发表于 2026-07-16
Hao Chen, Clara Grazian
摘要
Health technology assessment frequently requires survival predictions well beyond the observed trial follow-up, yet single parametric models fitted to short horizons can accumulate long-term bias. Current guidance also encourages the principled use of external evidence, such as registry summaries or expert anchors, while maintaining fidelity to the trial data. We introduce an adaptive spline-weighted blended extrapolation on the cumulative-hazard scale that unifies these aims. The observation side is fitted on the trial window using a piecewise-exponential Cox-type model with a smooth prior-driven continuation beyond the administrative cutoff, implemented via INLA. The external side is an anchored Gompertz tail identified by a prespecified survival level at a clinically relevant time. To blend the two components, we compare their cumulative hazards, learn a monotone P-spline score over time, and pass it through a logistic link to obtain a data-driven, time-varying weight. A simple “temperature” scaling controls the slope of this weight and provides a practical guarantee of non-negative blended hazards on a chosen grid, preserving adaptivity while ensuring feasibility. Across Monte Carlo scenarios spanning multiple tail shapes and censoring levels, the method delivers consistently lower absolute survival error, smaller restricted mean survival time error, and improved stability compared with fixed-schedule blending and single-family parametric models. In a SEER registry study with three-year observation and ten-year extrapolation, the blended curve tracks the Kaplan–Meier estimates within follow-up and transitions smoothly toward the anchored tail, yielding small long-horizon errors across cancer sites and age strata. The framework is modular, interpretable, and easily extended to alternative tails and multiple anchors. An open-source implementation is available in the R package survblendr at https://github.com/haohaostats/survblendr .
JASA Bayesian Methods · 发表于 2026-07-15,收录于 2026-07-24
Not available
摘要
We develop sampling methods, which consist of Gaussian invariant versions of random walk Metropolis (RWM), Metropolis adjusted Langevin algorithm (MALA) and second order Hessian or Manifold MALA. Unlike standard RWM and MALA, we show that Gaussian invariant sampling can lead to ergodic estimators with improved statistical efficiency. This is due to a remarkable property of Gaussian invariance that allows us to obtain exact analytical solutions to the Poisson equation for Gaussian targets. These solutions can be used to construct efficient and easy to use control variates for variance reduction of estimators under any intractable target. We demonstrate the new samplers and estimators in several examples, including high dimensional targets in latent Gaussian models where we compare against several advanced methods and obtain state-of-the-art results. We also provide theoretical results regarding geometric ergodicity, and an optimal scaling analysis that shows the dependence of the optimal acceptance rate on the Gaussianity of the target. *(abstract via OpenAlex)*
SMMR Survival Analysis · 发表于 2026-07-15
Hein Putter, Chengyuan Lu, Jacco Wallinga
摘要
Compartmental models based on ordinary differential equations quantifying the interactions between susceptible, infectious and recovered individuals within a population have played an important role in infectious disease modelling. The aim of the present paper is to explain the link between stochastic epidemic models based on the susceptible-infectious-recovered (SIR) model and methods from survival analysis. We illustrate how standard software for survival analysis in the statistical language R can be used to estimate pivotal parameters in the stochastic SIR model in the very much idealized situation where the epidemic is completely observed. Extensions incorporating interventions, age structure and heterogeneity are explored and illustrated.
SiM Clinical Trials Causal Inference · 发表于 2026-07-15
Ryo Emoto, Kiyoaki Ishii, Toshinari Takamura, Shigeyuki Matsui
摘要
Recent advances in biotechnology and personalized medicine have driven the development of efficient clinical trial methodologies for assessing treatment efficacy across multiple populations defined by treatment effect modifiers. Within‐patient comparison of different treatments is a promising approach for improving study efficiency across multiple populations by eliminating between‐patient variability in treatment evaluation. This study provides a framework for evaluating treatment efficacy in multiple populations for crossover trials and evaluates the efficacy gain in comparison with the standard parallel‐group analysis. Simulation experiments confirm that the crossover analysis consistently outperforms the parallel‐group analysis in statistical power, especially when carryover effects are small. An application to a clinical trial in Type 2 diabetes demonstrates the efficiency advantages of the crossover analysis. These numerical results emphasize the potential of the crossover analysis for enhancing the efficiency of clinical development of personalized medicine.
JASA High-Dimensional Econometrics & Economics · 发表于 2026-07-15,收录于 2026-06-05
Yifan Gu, Hanfang Yang, Songshan Yang, Hui Zou
摘要
In modern data analysis, an improvement in statistical efficiency is expected via effective collaboration among multiple data holders with non-shared data. In this article, we propose a collaborative score-type test (CST) for testing linear hypotheses, which accommodates potentially high-dimensional nuisance parameters and a diverging number of constraints and target parameters. Through a careful decomposition of the Kiefer-Bahadur representation for the traditional score statistic, we identify and approximate the key components using aggregated local gradient information from each data source. In addition, we employ a two-stage partial penalization strategy to shrink the approximation error and mitigate the bias from the high-dimensional nuisance parameters. Unlike existing methods, the CST procedure involves constrained optimization under non-shared and high-dimensional data settings, which requires novel theoretical developments. We derive the limiting distributions for the CST statistic under the null hypothesis and the local alternatives. Besides, the CST exhibits an oracle property and achieves the global statistical efficiency. Moreover, it relaxes the stringent restrictions on the number of data sources required in the current literature. Extensive numerical studies and a real-world example demonstrate the effectiveness and validity of our proposed method. *(abstract via OpenAlex)*
SMMR Causal Inference Bayesian Methods · 发表于 2026-07-15
Jinhong Cui, Xiaoxiao Zhou, Melissa J Smith, Andrew Sims, Justin M Leach, D. Leann Long, Nengjun Yi
摘要
Mediation analysis is a powerful tool for exploring the causal relationships between exposures and outcomes that are mediated by intermediate variables. In this paper, we propose a flexible Bayesian mediation analysis framework to accommodate zero-inflated mediators, compatible with a wide range of outcome distributions. This novel technique employs Bayesian models for both the mediator and the outcome, utilizing Markov Chain Monte Carlo algorithms for parameter estimation. While addressing the challenges posed by an excess of zeros, we further decompose the mediation effects into components influenced by either the probability of zero or the mean of the non-zero distribution in the mediator. An associated R package mediationBayes ( https://github.com/jhcuibst/mediationBayes.git ) has been developed to facilitate the application of this framework. Through comprehensive simulation studies, we demonstrate that our method outperforms alternatives in terms of the accuracy of point estimates, coverage probabilities, and the precision of the mediation effects decomposition. We further illustrate the practical applicability of our method by conducting an analysis on the REasons for Geographic And Racial Differences in Stroke Study to investigate the mediating influence of smoking pack-years on the association between educational levels and incident hypertension, where mediation effects are quantified on a risk ratio scale for binary outcomes.
SMMR Missing Data & Measurement Error · 发表于 2026-07-15
Tetiana Gorbach, Tim P Morris, James R Carpenter
摘要
While most methods for missing not at random (MNAR) data in regression models address MNAR outcomes assuming fully observed predictors, real-world observational health and longitudinal studies often violate this assumption. This paper compares several approaches for handling MNAR data in linear regression when missingness depends on both partially observed outcomes and predictors. Through extensive simulations, we evaluate complete-case analysis, multiple imputation assuming missing at random, maximum likelihood estimation via the Heckman selection model, uncertainty intervals, multiple imputation under the Heckman selection model, not-at-random fully conditional specification, imputation stacking, and random indicator imputation. None of the methods consistently produced unbiased estimates or nominal coverage across all scenarios. However, not-at-random fully conditional specification was straightforward to implement and yielded coverage close to the nominal level in most scenarios, provided that the sensitivity parameters were specified near their true values. Our results highlight the importance of sensitivity analyses exploring various full-data models and careful parameter specification when addressing MNAR in both outcomes and predictors. We illustrate such sensitivity analyses using Betula study data on the relationship between longitudinal memory change and grey matter volume in aging. The association remained significant across most considered MNAR scenarios, reinforcing existing evidence for this relationship.