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 22 页
共 1491 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2025-04-04 到 2025-03-22。
按期刊、按主题看 →
JASA High-Dimensional Latent Variables & Mixtures · 发表于 2025-04-04
Alexis Boulin, Elena Di Bernardino, Thomas Laloë, Gwladys Toulemonde
摘要
We propose a new class of models for variable clustering called Asymptotic Independent block (AI-block) models, which defines population-level clusters based on the independence of the maxima of a multivariate stationary mixing random process among clusters. This class of models is identifiable, meaning that there exists a maximal element with a partial order between partitions, allowing for statistical inference. We also present an algorithm depending on a tuning parameter that recovers the clusters of variables without specifying the number of clusters a priori. Our work provides some theoretical insights into the consistency of our algorithm, demonstrating that under certain conditions it can effectively identify clusters in the data with a computational complexity that is polynomial in the dimension. A data-driven selection method for the tuning parameter is also proposed. To further illustrate the significance of our work, we applied our method to neuroscience and environmental real-datasets. These applications highlight the potential and versatility of the proposed approach. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work. *(abstract via OpenAlex)*
JASA Bayesian Methods Causal Inference · 发表于 2025-04-04
Jacob R. Englert, Stefanie T. Ebelt, Howard H. Chang
摘要
Epidemiological approaches for examining human health responses to environmental exposures in observational studies often control for confounding by implementing clever matching schemes and using statistical methods based on conditional likelihood. Nonparametric regression models have surged in popularity in recent years as a tool for estimating individual-level heterogeneous effects, which provide a more detailed picture of the exposure-response relationship but can also be aggregated to obtain improved marginal estimates at the population level. In this work we incorporate Bayesian additive regression trees (BART) into the conditional logistic regression model to identify heterogeneous exposure effects in a case-crossover design. Conditional logistic BART (CL-BART) utilizes reversible jump Markov chain Monte Carlo to bypass the conditional conjugacy requirement of the original BART algorithm. Our work is motivated by the growing interest in identifying subpopulations more vulnerable to environmental exposures. We apply CL-BART to a study of the impact of heat waves on people with Alzheimer's disease in California and effect modification by other chronic conditions. Through this application, we also describe strategies to examine heterogeneous odds ratios through variable importance, partial dependence, and lower-dimensional summaries. *(abstract via PubMed)*
JASA Nonparametric & Semiparametric Bayesian Methods · 发表于 2025-04-03
Daniel R. Kowal, Bohan Wu
摘要
Data transformations are essential for broad applicability of parametric regression models. However, for Bayesian analysis, joint inference of the transformation and model parameters typically involves restrictive parametric transformations or nonparametric representations that are computationally inefficient and cumbersome for implementation and theoretical analysis, which limits their usability in practice. This paper introduces a simple, general, and efficient strategy for joint posterior inference of an unknown transformation and all regression model parameters. The proposed approach directly targets the posterior distribution of the transformation by linking it with the marginal distributions of the independent and dependent variables, and then deploys a Bayesian nonparametric model via the Bayesian bootstrap. Crucially, this approach delivers (1) joint posterior consistency under general conditions, including multiple model misspecifications, and (2) efficient Monte Carlo (not Markov chain Monte Carlo) inference for the transformation and all parameters for important special cases. These tools apply across a variety of data domains, including real-valued, positive, and compactly-supported data. Simulation studies and an empirical application demonstrate the effectiveness and efficiency of this strategy for semiparametric Bayesian analysis with linear models, quantile regression, and Gaussian processes. The R package SeBR is available on CRAN. *(abstract via PubMed)*
Biometrics High-Dimensional Genetics & Omics Testing & Uncertainty · 发表于 2025-04-02
Danning Li, Lingzhou Xue, Haoyi Yang, Xiufan Yu
摘要
Testing differences in mean vectors is a fundamental task in the analysis of high-dimensional microbiome compositional data. Existing methods may suffer from low power if the underlying signal pattern is in a situation that does not favor the deployed test. In this work, we develop 2-sample power-enhanced mean tests for high-dimensional compositional data based on the combination of $P$-values, which integrates strengths from 2 popular types of tests: the maximum-type test and the quadratic-type test. We provide rigorous theoretical guarantees on the proposed tests, showing accurate Type-I error rate control and enhanced testing power. Our method boosts the testing power toward a broader alternative space, which yields robust performance across a wide range of signal pattern settings. Our methodology and theory also contribute to the literature on power enhancement and Gaussian approximation for high-dimensional hypothesis testing. We demonstrate the performance of our method on both simulated data and real-world microbiome data, showing that our proposed approach improves the testing power substantially compared to existing methods.
AoS Other · 发表于 2025-04-01
Jean-Marie Dufour, Eric Renault, Victoria Zinde-Walsh
摘要
This paper provides an exhaustive characterization of the asymptotic null distribution of Wald-type statistics for testing restrictions given by polynomial functions—which may involve singularities—when the limiting distribution of the parameter estimator is absolutely continuous (e.g., Gaussian). In addition to the well-known finite-sample noninvariance, there is also an asymptotic noninvariance (nonpivotality): with standard critical values, the test may either under-reject or over-reject, and even diverge under the null hypothesis. The asymptotic distribution of the test statistic can vary under the null hypothesis and depends on the true unknown parameter value. All these situations are possible in testing restrictions which arise in the statistical and econometric literatures, for example, for examining the specification of ARMA models, causality at different horizons, indirect effects, zero determinant hypotheses on matrices of coefficients, and other situations where singularities cannot be excluded. We provide the limit distribution and general bounds for the single restriction case. For multiple restrictions, we give a necessary and sufficient condition for the existence of a limit distribution, and its form if it exists.
AoS Optimization & Learning Theory High-Dimensional · 发表于 2025-04-01
Andrea Montanari, Feng Ruan, Youngtak Sohn, Jun Yan
摘要
Modern machine learning classifiers often exhibit vanishing classification error on the training set. They achieve this by learning nonlinear representations of the inputs that map the data into linearly separable classes. Motivated by these phenomena, we revisit high-dimensional maximum margin classification for linearly separable data. We consider a stylized setting in which data (yi,xi), i≤n are i.i.d. with xi∼N(0,Σ) a p-dimensional Gaussian feature vector, and yi∈{+1,−1} a label whose distribution depends on a linear combination of the covariates ⟨θ∗,xi⟩. Recent universality results can be used to show that the results derived in the Gaussian setting also apply when xi=ϕ(zi) for standard Gaussian zi and ϕ a nonlinear featurization map. We consider the proportional asymptotics n,p→∞ with p/n→ψ, and derive exact expressions for the limiting generalization error. We use this theory to derive two results of independent interest: (i) Sufficient conditions on (Σ,θ∗) for “benign overfitting” that are parallel to previously derived conditions in the case of linear regression; (ii) An asymptotically exact expression for the generalization error when max-margin classification is used in conjunction with feature vectors produced by random one-layer neural networks.
AoS Other · 发表于 2025-04-01
Zhe Gao, Roulin Wang, Xueqin Wang, Heping Zhang
摘要
The exploration of associations between random objects with complex geometric structures has catalyzed the development of various novel statistical tests encompassing distance-based and kernel-based statistics. These methods have various strengths and limitations. One problem is that their test statistics tend to converge to asymptotic null distributions involving second-order Wiener chaos, which are hard to compute and need approximation or permutation techniques that use much computing power to build rejection regions. In this work, we take an entirely different and novel strategy by using the so-called “random-lifter.” This method is engineered to yield test statistics with the standard normal limit under null distributions without the need for sample splitting. In other words, we set our sights on having simple limiting distributions and finding the proper statistics through reverse engineering. We use the Central Limit Theorems (CLTs) for degenerate U-statistics derived from our novel association measures to do this. As a result, the asymptotic distributions of our proposed tests are straightforward to compute. Our test statistics also have the minimax property. We further substantiate that our method maintains competitive power against existing methods with minimal adjustments to constant factors. Both numerical simulations and real-data analysis corroborate the efficacy of the random-lifter method.
AoS Nonparametric & Semiparametric High-Dimensional · 发表于 2025-04-01
Ivo V. Stoepker, Rui M. Castro, Ery Arias-Castro
摘要
Detecting anomalies in large sets of observations is crucial in various applications, such as epidemiological studies, gene expression studies, and systems monitoring. We consider settings where the units of interest result in multiple independent observations from potentially distinct referentials. Scan statistics and related methods are commonly used in such settings, but rely on stringent modeling assumptions for proper calibration. We instead propose a rank-based variant of the higher criticism statistic that only requires independent observations originating from ordered spaces. We show under what conditions the resulting methodology is able to detect the presence of anomalies. These conditions are stated in a general, nonparametric manner, and depend solely on the probabilities of anomalous observations exceeding nominal observations. The analysis requires a refined understanding of the distribution of the ranks under the presence of anomalies, and in particular of the rank-induced dependencies. The methodology is robust against heavy-tailed distributions through the use of ranks. Within the exponential family and a family of convolutional models, we analytically quantify the asymptotic performance of our methodology and the performance of the oracle, and show the difference is small for many common models. Simulations confirm these results. We show the applicability of the methodology through an analysis of quality control data of a pharmaceutical manufacturing process.
AoS Testing & Uncertainty · 发表于 2025-04-01
Kaiyue Wen, Tengyao Wang, Yuhao Wang
摘要
We consider the problem of testing whether a single coefficient is equal to zero in linear models when the dimension of covariates p can be up to a constant fraction of sample size n. In this regime, an important topic is to propose tests with finite-sample valid size control without requiring the noise to follow strong distributional assumptions. In this paper, we propose a new method, called the residual permutation test (RPT), which is constructed by projecting the regression residuals onto the space orthogonal to the union of the column spaces of the original and permuted design matrices. RPT can be proved to achieve finite-sample size validity under fixed design with just exchangeable noises, whenever p<n/2. Moreover, RPT is shown to be asymptotically powerful for heavy-tailed noises with bounded (1+t)th order moment when the true coefficient is at least of order n−t/(1+t) for t∈[0,1]. We further proved that this signal size requirement is essentially rate-optimal in the minimax sense. Numerical studies confirm that RPT performs well in a wide range of simulation settings with normal and heavy-tailed noise distributions.
AoS Other · 发表于 2025-04-01
Ankur Moitra, Alexander S. Wein
摘要
We revisit the fundamental question of simple-versus-simple hypothesis testing with an eye toward computational complexity, as the statistically optimal likelihood ratio test is often computationally intractable in high-dimensional settings. In the classical spiked Wigner model with a general i.i.d. spike prior, we show (conditional on a conjecture) that an existing test based on linear spectral statistics achieves the best possible trade-off curve between type-I and type-II error rates among all computationally efficient tests, even though there are exponential-time tests that do better. This result is conditional on an appropriate complexity-theoretic conjecture, namely a natural strengthening of the well-established low-degree conjecture. Our result shows that the spectrum is a sufficient statistic for computationally bounded tests (but not for all tests). To our knowledge, our approach gives the first tool for reasoning about the precise asymptotic testing error achievable with efficient computation. The main ingredients required for our hardness result are a sharp bound on the norm of the low-degree likelihood ratio along with (counterintuitively) a positive result on achievability of testing. This strategy appears to be new even in the setting of unbounded computation, in which case it gives an alternate way to analyze the fundamental statistical limits of testing.
AoS Other · 发表于 2025-04-01
Pierre C. Bellec
摘要
We consider observations (X,y) from single index models with unknown link function, Gaussian covariates and a regularized M-estimator βˆ constructed from convex loss function and regularizer. In the regime where sample size n and dimension p are both increasing such that p/n has a finite limit, the behavior of the empirical distribution of βˆ and the predicted values Xβˆ has been previously characterized in a number of models: The empirical distributions are known to converge to proximal operators of the loss and penalty in a related Gaussian sequence model, which captures the interplay between ratio p n, loss, regularization and the data generating process. This connection between (βˆ,Xβˆ) and the corresponding proximal operators involves mean-field parameters defined as solutions to a nonlinear system of equations. This system typically involve unobservable quantities such as the prior distribution on the index or the link function, so the mean-field parameters need to be estimated. Although estimators for the mean-field parameters have been proposed in specific cases, a general framework that applies simultaneously to a broad class of loss and penalty has so far been missing. This paper develops a different theory to describe the empirical distribution of βˆ and Xβˆ: Approximations of (βˆ,Xβˆ) in terms of proximal operators are provided that only involve observable adjustments in place of the mean-field parameters. These proposed observable adjustments are data-driven, for example, do not require prior knowledge of the index or the link function. These new adjustments yield confidence intervals for individual components of the index, as well as estimators of the correlation of βˆ with the index, enabling parameter tuning to maximize the correlation. The interplay between loss, regularization and the model is captured in a data-driven manner, without relying on the nonlinear systems studied in previous works. The results are proved to hold both strongly convex regularizers and unregularized M-estimation.
AoS Testing & Uncertainty · 发表于 2025-04-01
Dmitriy Kunisky
摘要
We study when low coordinate degree functions (LCDF)—linear combinations of functions depending on small subsets of entries of a vector—can hypothesis test between high-dimensional probability measures. These functions are a generalization, proposed in Hopkins’ 2018 thesis but seldom studied since, of low degree polynomials (LDP), a class widely used in recent literature as a proxy for all efficient algorithms for tasks in statistics and optimization. Instead of the orthogonal polynomial decompositions used in LDP calculations, our analysis of LCDF is based on the Efron–Stein or ANOVA decomposition, making it much more broadly applicable. By way of illustration, we prove channel universality for the success of LCDF in testing for the presence of sufficiently “dilute” random signals through noisy channels: the efficacy of LCDF depends on the channel only through the scalar Fisher information for a class of channels including nearly arbitrary additive i.i.d. noise and nearly arbitrary exponential families. As applications, we extend lower bounds against LDP for spiked matrix and tensor models under additive Gaussian noise to lower bounds against LCDF under general noisy channels. We also give a simple and unified treatment of the effect of censoring models by erasing observations at random and of quantizing models by taking the sign of the observations. These results are the first computational lower bounds against any large class of algorithms for all of these models when the channel is not one of a few special cases, and thereby give the first substantial evidence for the universality of several statistical-to-computational gaps.
AoS Optimization & Learning Theory · 发表于 2025-04-01
Arnab Auddy, Ming Yuan
摘要
In this paper, we investigate the optimal statistical performance and the impact of computational constraints for independent component analysis (ICA). Our goal is twofold. On the one hand, we characterize the precise role of dimensionality on sample complexity and statistical accuracy, and how computational consideration may affect them. In particular, we show that the optimal sample complexity is linear in dimensionality, and interestingly, the commonly used sample kurtosis-based approaches are necessarily suboptimal. However, the optimal sample complexity becomes quadratic, up to a logarithmic factor, in the dimension if we restrict ourselves to estimates that can be computed with low-degree polynomial algorithms. On the other hand, we develop computationally tractable estimates that attain both the optimal sample complexity and minimax optimal rates of convergence. We study the asymptotic properties of the proposed estimates and establish their asymptotic normality that can be readily used for statistical inferences. Our method is fairly easy to implement and numerical experiments are presented to further demonstrate its practical merits.
AoS Other · 发表于 2025-04-01
Lei Shi, Jingshen Wang, Peng Ding
摘要
Ever since the seminal work of R. A. Fisher and F. Yates, factorial designs have been an important experimental tool to simultaneously estimate the effects of multiple treatment factors. In factorial designs, the number of treatment combinations grows exponentially with the number of treatment factors, which motivates the forward selection strategy based on the sparsity, hierarchy and heredity principles for factorial effects. Although this strategy is intuitive and has been widely used in practice, its rigorous statistical theory has not been formally established. To fill this gap, we establish design-based theory for forward factor selection in factorial designs based on the potential outcome framework. We not only prove a consistency property for the factor selection procedure but also discuss statistical inference after factor selection. In particular, with selection consistency, we quantify the advantages of forward selection based on asymptotic efficiency gain in estimating factorial effects. With inconsistent selection in higher-order interactions, we propose two strategies and investigate their impact on subsequent inference. Our formulation differs from the existing literature on variable selection and post-selection inference because our theory is based solely on the physical randomization of the factorial design and does not rely on a correctly specified outcome model.
AoS Other · 发表于 2025-04-01
Ery Arias-Castro, Wanli Qiao
摘要
We adapt concepts, methodology, and theory originally developed in the areas of multidimensional scaling and dimensionality reduction for Euclidean data to be applicable to distributional data. We focus on classical scaling and Isomap—prototypical methods that have played important roles in these areas—and showcase their use in the context of distributional data analysis. In the process, we highlight the crucial role that the ambient metric plays.
AoS Other · 发表于 2025-04-01
Yuling Jiao, Lican Kang, Jin Liu, Xiliang Lu, Jerry Zhijian Yang
摘要
In this paper, we consider deep approximate policy iteration (DAPI) with the Bellman residual minimization in reinforcement learning. In each iteration of DAPI, we apply convolutional neural networks (CNNs) with ReLU activation, called ReLU CNNs, to estimate the fixed point of the Bellman equation by minimizing an unbiased minimax loss. To bound the estimation error in each iteration, we control the statistical and approximation errors using the tools of the empirical process theory with dependent data and deep approximation theory, respectively. We establish a novel statistical error bound for ReLU CNNs on dependent data that is C-mixing, and an approximation error bound for ReLU CNNs on Hölder class. Combining with error propagation, we obtain a nonasymptotic error bound between the optimal action-value function Q∗ and the estimated Q function induced by the greedy policy in DAPI. This bound depends on the sample size and ambient dimension of the data, as well as the size, weight bound, and depth of the CNNs, providing prior guidance on how to set these hyperparameters to achieve the desired convergence rate when training DAPI in practice. Moreover, this bound circumvents the curse of dimensionality if the distribution of state-action pairs is supported on a set with a low intrinsic dimension.
AoS Other · 发表于 2025-04-01
Cristina Butucea, Alexander Meister, Angelika Rohde
摘要
We consider a general class of statistical experiments, in which an n-dimensional centered Gaussian random variable is observed and its covariance matrix is the parameter of interest. The covariance matrix is assumed to be well-approximable in a linear space of lower dimension Kn with eigenvalues uniformly bounded away from zero and infinity. We prove asymptotic equivalence of this experiment and a class of Kn-dimensional Gaussian models with informative expectation in Le Cam’s sense when n tends to infinity and Kn is allowed to increase moderately in n at a polynomial rate. For this purpose we derive a new localization technique for non-i.i.d. data and a novel high-dimensional Central Limit Law in total variation distance. These results are key ingredients to show asymptotic equivalence between the experiments of locally stationary Gaussian time series and a bivariate Wiener process with the log spectral density as its drift. On this way a novel class of matrices is introduced which generalizes circulant Toeplitz matrices traditionally used for strictly stationary time series.
AoS Machine Learning · 发表于 2025-04-01
Junlong Zhao, Xiumin Liu, Bin Du, Yufeng Liu
摘要
Converting a continuous variable into a discrete one is a commonly used technique for solving various problems in both statistics and machine learning. It is well known that discretizations result in biases. However, this issue has not been studied systematically. In this paper, a general framework is proposed to understand and compare the approximation errors of different slicing strategies. Poincaré-type inequalities are first established for univariate discretizations and then generalized to the multivariate and other settings. It is shown that the bias is controlled by two factors: the distance between two specific distributions that are generated with and without discretizations respectively, and the smoothness of the functions involved. Several important applications are considered to illustrate the usefulness of the results. Our results help to understand the approximation error of some matrix used in the literature of dimension reduction. Furthermore, as an illustration of the usefulness of discretizations, we propose an algorithm for regression problems, by combining random forest with partial discretizations of responses. Simulation results confirm the advantages of this algorithm over the classical random forest.
AoS Testing & Uncertainty · 发表于 2025-04-01
Yingying Fan, Lan Gao, Jinchi Lv
摘要
We investigate the robustness of the model-X knockoffs framework with respect to the misspecified or estimated feature distribution. We achieve such a goal by theoretically studying the feature selection performance of a practically implemented knockoffs algorithm, which we name as the approximate knockoffs (ARK) procedure, under the measures of the false discovery rate (FDR) and k-familywise error rate (k-FWER). The approximate knockoffs procedure differs from the model-X knockoffs procedure only in that the former uses the misspecified or estimated feature distribution. A key technique in our theoretical analyses is to couple the approximate knockoffs procedure with the model-X knockoffs procedure so that random variables in these two procedures can be close in realizations. We prove that if such coupled model-X knockoffs procedure exists, the approximate knockoffs procedure can achieve the asymptotic FDR or k-FWER control at the target level. We showcase three specific constructions of such coupled model-X knockoff variables, verifying their existence and justifying the robustness of the model-X knockoffs framework. Additionally, we formally connect our concept of knockoff variable coupling to a type of Wasserstein distance.
AoS Spatial & Time Series Testing & Uncertainty · 发表于 2025-04-01
Anne van Delft, Andrew J. Blumberg
摘要
We introduce a new framework to analyze shape descriptors that capture the geometric features of an ensemble of point clouds. At the core of our approach is the point of view that the data arises as sampled recordings from a metric space-valued stochastic process, possibly of nonstationary nature, thereby integrating geometric data analysis into the realm of functional time series analysis. Our framework allows for natural incorporation of spatial-temporal dynamics, heterogeneous sampling, and the study of convergence rates. Further, we derive complete invariants for classes of metric space-valued stochastic processes in the spirit of Gromov, and relate these invariants to so-called ball volume processes. Under mild dependence conditions, a weak invariance principle in D([0,1]×[0,R]) is established for sequential empirical versions of the latter, assuming the probabilistic structure possibly changes over time. Finally, we use this result to introduce novel test statistics for topological change, which are distribution-free in the limit under the hypothesis of stationarity. We explore these test statistics on time series of single-cell mRNA expression data, using shape descriptors coming from topological data analysis.
AoS Econometrics & Economics High-Dimensional Causal Inference · 发表于 2025-04-01
Kuanhao Jiang, Rajarshi Mukherjee, Subhabrata Sen, Pragya Sur
摘要
Estimation of the average treatment effect (ATE) is a central problem in causal inference. In recent times, inference for the ATE in the presence of high-dimensional covariates has been extensively studied. Among diverse approaches that have been proposed, augmented inverse propensity weighting (AIPW) with cross-fitting has emerged a popular choice in practice. In this work, we study this cross-fit AIPW estimator under well-specified outcome regression and propensity score models in a high-dimensional regime where the number of features and samples are both large and comparable. Under assumptions on the covariate distribution, we establish a new central limit theorem for the suitably scaled cross-fit AIPW that applies without any sparsity assumptions on the underlying high-dimensional parameters. Our CLT uncovers two crucial phenomena among others: (i) the AIPW exhibits a substantial variance inflation that can be precisely quantified in terms of the signal-to-noise ratio and other problem parameters, (ii) the asymptotic covariance between the precross-fit estimators is nonnegligible even on the n scale. These findings are strikingly different from their classical counterparts. On the technical front, our work utilizes a novel interplay between three distinct tools—approximate message passing theory, the theory of deterministic equivalents and the leave-one-out approach. We believe our proof techniques should be useful for analyzing other two-stage estimators in this high-dimensional regime. We complement our theoretical results with simulations that demonstrate both the finite sample efficacy of our CLT and its robustness to our assumptions. Finally, we provide some theoretical evidence for the universality of our CLT to the law of the covariates, and explore the effects of certain forms of model misspecification.
SMMR Other · 发表于 2025-03-31
Baoshan Zhang, Yuan Wu
摘要
This study addresses a critical gap in the design of clinical trials that use grouped sequential designs for one-sample or paired ordinal categorical outcomes. Single-arm experiments, such as those using the modified Rankin Scale in stroke trials, underscore the necessity of our work. We present a novel method for applying the Wilcoxon signed-rank test to grouped sequences in these contexts. Our approach provides a practical and theoretical framework for assessing treatment effects, detailing variance formulas and demonstrating the asymptotic normality of the U-statistic. Through simulation studies and real data analysis, we validate the empirical Type I error rates and power. Additionally, we include a comprehensive flowchart to guide researchers in determining the required sample size to achieve specified power levels while controlling Type I error rates, thereby enhancing the design process of sequential trials.
SMMR Bayesian Methods · 发表于 2025-03-31
Duncan T Wilson, Andrew Hall, Julia M Brown, Rebecca EA Walwyn
摘要
Pilot trials are often conducted in advance of definitive trials to assess their feasibility and to inform their design. Although pilot trials typically collect primary endpoint data, preliminary tests of effectiveness have been discouraged given their typically low power. Power could be increased at the cost of a higher type I error rate, but there is little methodological guidance on how to determine the optimal balance between these operating characteristics. We consider a Bayesian decision-theoretic approach to this problem, introducing a utility function and defining an optimal pilot and definitive trial programme as that which maximises expected utility. We base utility on changes in average primary outcome, the cost of sampling, treatment costs, and the decision-maker’s attitude to risk. We apply this approach to re-design OK-Diabetes, a pilot trial of a complex intervention with a continuous primary outcome with known standard deviation. We then examine how optimal programme characteristics vary with the parameters of the utility function. We find that the conventional approach of not testing for effectiveness in pilot trials can be considerably sub-optimal.
SMMR Other · 发表于 2025-03-31
Jiahui Xin, Wei Ma
摘要
In the context of precision medicine, covariate-adjusted response-adaptive (CARA) randomization has garnered much attention from both academia and industry due to its benefits in providing ethical and tailored treatment assignments based on patients’ profiles while still preserving favorable statistical properties. Recent years have seen substantial progress in inference for various adaptive experimental designs. In particular, research has focused on two important perspectives: how to obtain robust inference in the presence of model misspecification, and what the smallest variance, i.e., the efficiency bound, an estimator can achieve. Notably, Armstrong (2022) derived the asymptotic efficiency bound for any randomization procedure that assigns treatments depending on covariates and accrued responses, thus including CARA, among others. However, to the best of our knowledge, no existing literature has addressed whether and how this bound can be achieved under CARA. In this paper, by connecting two strands of adaptive randomization literature, namely robust inference and efficiency bound, we provide a definitive answer in an important practical scenario where only discrete covariates are observed and used for stratification. We consider a special type of CARA, i.e., a stratified version of doubly-adaptive biased coin design and prove that the stratified difference-in-means estimator achieves Armstrong (2022)’s efficiency bound, with possible ethical constraints on treatment assignments. Our work provides new insights and demonstrates the potential for more research on CARA designs that maximize efficiency while adhering to ethical considerations. Future studies could explore achieving the asymptotic efficiency bound for CARA with continuous covariates, which remains an open question.
Biometrics Nonparametric & Semiparametric Latent Variables & Mixtures Bayesian Methods · 发表于 2025-03-31
Li-Pang Chen
摘要
Mixture models have been an important topic in statistical analysis, aiming to describe heterogeneous and complex data. A large body of estimation methods and developments have been discussed over the decades; however, few monographs provide a comprehensive and systematic summary. This review introduces the 2024 book Mixture Models: Parametric, Semiparametric, and New Directions by Yao and Xiang. While there are some monographs that present mixture models (eg, McLachlan and Peel 2000; Frühwirth-Schnatter, 2006; Frühwirth-Schnatter et al., 2019) with some overlapped materials, this book effectively summarizes basic ideas of mixture models and presents state-of-the-art developments that have not been discussed in the existing monographs; some detailed comparisons are deferred to the following paragraphs. Moreover, the authors provide theoretical results with rigorous proofs, along with the R programming code for computational implementation and applications for real datasets. This book contains 10 chapters. The authors first provide a big picture of mixture models and summarize basic frameworks in Chapter 1, including the (nonparametric) maximum-likelihood estimation, the Expectation-Maximization (EM) algorithm, mixture of normal/skewed distributions, and semi-supervised mixture models. Those materials generally cover various chapters in existing monographs, such as normal mixtures in McLachlan and Peel (2000, Chapter 3) and Frühwirth-Schnatter (2006, Chapter 6), and non-normal mixtures in McLachlan and Peel (2000, Chapter 5) and Frühwirth-Schnatter (2006, Chapter 9). In contrast to the discussion of continuous random variables, Chapter 2 focuses on discrete data, including categorical and counting data. The authors primarily use hidden Markov models and latent class models to characterize discrete data. Additionally, mixed data are discussed within the context of mixture models. Beginning in Chapter 3, the authors present mixture models under various regression structures. Specifically, Chapter 3 introduces various types of mixture regression models, including mixtures of (generalized) linear models, expert models, hidden Markov models, and mixed-effects models. In contrast to frequentist inference, Chapter 4 discusses Bayesian mixture models. In addition to standard Bayesian methods for mixture analysis, the authors also discuss Bayesian normal mixture models. Generally, the fundamental introduction in this chapter is almost the same as McLachlan and Peel (2000, Chapter 4) and Frühwirth-Schnatter (2006, Chapter 5), but the authors additionally discuss the multivariate mixtures that were not explored before. Chapter 5 focuses the discussion on label switching for mixture models, which is caused by relabeling the mixture components. To address the label-switching problem, the authors propose 4 different methods in each section separately, including loss-functions-based methods, modal-relabeling methods, soft probabilistic relabeling methods, and label-switching strategy based on the frequentist perspective. Regression analysis involves several crucial issues, such as model fit and selection, estimation with high-dimensional data, and semiparametric regression by relaxing restrictive parametric constraints. The authors extend those statistical problems to handle mixture models. For example, Chapter 6 introduces the likelihood ratio test for mixture models to do hypothesis test, which overlaps the presentation in Frühwirth-Schnatter et al. (2019, Chapter 7). In addition, the new material in this book includes more techniques, such as penalized mixture models and EM test for model selection. Chapter 7 extends an idea of robust regression to examine robust mixture regression models. Unlike existing approaches, such as Chapter 8 in Frühwirth-Schnatter (2006), which simply considers a regression approach, this chapter proposes a modified EM algorithm for robust estimation in complex settings, such as regression models with heavy-tailed error, log-concave regression, or cluster-weighted modeling. Chapter 8 considers mixture models subject to high-dimensional data, and several methods are discussed in each section separately. While Frühwirth-Schnatter et al. (2019, Chapter 11) also discussed this topic, they mainly focused on the introduction of several types of data; in contrast, this book summarizes several estimation procedures. Those methods include factor analyzers, reduced projections, regularized mixture modeling, random projection, and penalized estimation function approaches. Finally, Chapters 9 and 10 study semiparametric mixture (regression) models that are innovative in this book and have not been well summarized in the existing monographs (eg, Frühwirth-Schnatter, 2006; McLachlan and Peel, 2000; Frühwirth-Schnatter et al., 2019). In Chapter 9, the authors focus on the estimations for various settings, including location-shifted models, models with shape constraints, multivariate mixture models, hidden Markov models, and Bayesian nonparametric mixture models; on the other hand, with the concept of “regression” accommodated, Chapter 10 discusses several regression structures, including nonparametric regression models, regression models with functional/longitudinal data, and machine learning based regression models. After reading this book, I think the authors well summarize the frameworks of mixture models, including fundamental concepts and modern developments for complex data structures. The general presentation is a logical progression and the materials in the chapters are consecutive, from the fundamental concepts to advanced and new development. Rigorous mathematical derivations, detailed pseudocode, and computational implementations for numerical studies show the balance between methodologies and applications, and offer readers a comprehensive view, making the content easy to follow and the insights clear. In general, this book is a good reference for researchers, data scientists, and professionals whose work is about mixture models since it contains many practical R code demonstrations for real-world applications, together with a clear interpretation for real datasets. In addition, this book is recommended to be a suitable textbook for newer or students who are learning mixture models because the authors present all the materials and mathematical formulations in a clear style. None declared. *(abstract via OpenAlex)*
SMMR Networks & Graphs Diagnostics & Prediction Survey & Sampling · 发表于 2025-03-31
A James O’Malley, Yifan Zhao, Carly Bobak, Chuanling Qin, Erika L Moen, Daniel N Rockmore
摘要
There is growing use of shared-patient physician networks in health services research and practice, but minimal study of the consequences of decisions made in constructing them. To address this gap, we surveyed physician employees of a National Physician Organization (NPO) on their peer physician relationships. Using the physicians’ survey nominations as ground truths, we evaluated the diagnostic accuracy of shared-patient edge-weights and the optimal construction of physician networks from sequences of patient-physician encounters. To further improve diagnostic accuracy, we optimized network construction with respect to the within-dyad difference and summation of edge-strength (two orthogonal measures), optimally combining them to form a final edge-weight. To achieve these goals, we develop statistical procedures to quantify the extent that directionality and other features of referral paths yield edge-weights with improved diagnostic properties. We also develop network models of the survey nominations incorporating directed (edge) and undirected (dyadic) shared-patient network measures as edge and dyad attributes to demonstrate that the measurement of the network as a whole is improved. Finally, we estimate the association of the physicians’ centrality in the NPO shared-patient network (a sociocentric feature that cannot be evaluated for the partially-measured survey-based network) with their beliefs regarding physician peer-influence.
SMMR Missing Data & Measurement Error Causal Inference · 发表于 2025-03-31
Jonathan W Bartlett, Camila Olarte Parra, Emily Granger, Ruth H Keogh, Erik W van Zwet, Rhian M Daniel
摘要
G-formula is a popular approach for estimating the effects of time-varying treatments or exposures from longitudinal data. G-formula is typically implemented using Monte-Carlo simulation, with non-parametric bootstrapping used for inference. In longitudinal data settings missing data are a common issue, which are often handled using multiple imputation, but it is unclear how G-formula and multiple imputation should be combined. We show how G-formula can be implemented using Bayesian multiple imputation methods for synthetic data, and that by doing so, we can impute missing data and simulate the counterfactuals of interest within a single coherent approach. We describe how this can be achieved using standard multiple imputation software and explore its performance using a simulation study and an application from cystic fibrosis.
SMMR Survival Analysis Clinical Trials Nonparametric & Semiparametric · 发表于 2025-03-31
Yi-Cheng Tai, Weijing Wang, Martin T Wells
摘要
In this article, we develop nonparametric inference methods for comparing survival data across two samples, beneficial for clinical trials of novel cancer therapies where long-term survival is critical. These therapies, including immunotherapies and other advanced treatments, aim to establish durable effects. They often exhibit distinct survival patterns such as crossing or delayed separation and potentially leveling-off at the tails of survival curves, violating the proportional hazards assumption and rendering the hazard ratio inappropriate for measuring treatment effects. Our methodology uses the mixture cure framework to separately analyze cure rates of long-term survivors and the survival functions of susceptible individuals. We evaluated a nonparametric estimator for the susceptible survival function in a one-sample setting. Under sufficient follow-up, it is expressed as a location-scale-shift variant of the Kaplan-Meier estimator. It retains desirable features of the Kaplan-Meier estimator, including inverse-probability-censoring weighting, product-limit estimation, self-consistency, and nonparametric efficiency. Under insufficient follow-up, it can be adapted by incorporating a suitable cure rate estimator. In the two-sample setting, in addition to using the difference in cure rates to measure long-term effects, we propose a graphical estimand to compare relative treatment effects on susceptible subgroups. This process, inspired by Kendall’s tau, compares the order of survival times among susceptible individuals. Large-sample properties of the proposed methods are derived for inference and their finite-sample properties are evaluated through simulations. The methodology is applied to analyze digitized data from the CheckMate 067 trial.
SMMR Clinical Trials Econometrics & Economics Bayesian Methods · 发表于 2025-03-31
Sydney Porter, Anne Eaton, Thomas A Murray
摘要
Recently, targeted and immunotherapy cancer treatments have motivated dose-finding based on efficacy-toxicity trade-offs rather than toxicity alone. The EffTox and utility-based Bayesian optimal interval (U-BOIN) dose-finding designs were developed in response to this need, but may be sensitive to elicited subjective design parameters that reflect the trade-off between efficacy and toxicity. To ease elicitation and reduce subjectivity, we propose dose desirability criteria that only depend on a preferential ordering of the joint efficacy-toxicity outcomes. We propose two novel order-based criteria and compare them with utility-based and contour-based criteria when paired with the design framework and probability models of EffTox and U-BOIN. The proposed dose desirability criteria simplify implementation and improve robustness to the elicited subjective design parameters and perform similarly in simulation studies to the established EffTox and U-BOIN designs when the ordering of the joint outcomes is equivalent. We also propose an alternative dose admissibility criteria based on the joint efficacy and toxicity profile of a dose rather than its marginal toxicity and efficacy profile. We argue that this alternative joint criterion is more consistent with defining dose desirability in terms of efficacy-toxicity trade-offs than the standard marginal admissibility criteria. The proposed methods enhance the usability and robustness of dose-finding designs that account for efficacy-toxicity trade-offs to identify the optimal biological dose.
SMMR Diagnostics & Prediction · 发表于 2025-03-31
Alessandro Baldi Antognini, Sara Cecconi, Rosamarie Frieri, Maroussa Zagoraiou
摘要
The rapidly developing field of personalized medicine is giving the opportunity to treat patients with a specific regimen according to their individual demographic, biological, or genomic characteristics, known also as biomarkers. While binary biomarkers simplify subgroup selection, challenges arise in the presence of continuous ones, which are often categorized based on data-driven quantiles. In the context of binary response trials for treatment comparisons, this paper proposes a method for determining the optimal cutoff of a continuous predictive biomarker to discriminate between sensitive and insensitive patients, based on their relative risk. We derived the optimal design to estimate such a cutoff, which requires a set of equality constraints that involve the unknown model parameters and the patients’ biomarker values and are not directly attainable. To implement the optimal design, a novel covariate-adjusted response-adaptive randomization is introduced, aimed at sequentially minimizing the Euclidean distance between the current allocation and the optimum. An extensive simulation study shows the performance of the proposed approach in terms of estimation efficiency and variance of the estimated cutoff. Finally, we show the potential severe ethical impact of adopting the data-dependent median to identify the subpopulations.
SMMR Testing & Uncertainty High-Dimensional · 发表于 2025-03-31
Marco Molinari, Magne Thoresen
摘要
Simultaneously performing variable selection and inference in high-dimensional regression models is an open challenge in statistics and machine learning. The increasing availability of vast amounts of variables requires the adoption of specific statistical procedures to accurately select the most important predictors in a high-dimensional space, while controlling the false discovery rate (FDR) associated with the variable selection procedure. In this paper, we propose the joint adoption of the Mirror Statistic approach to FDR control, coupled with outcome randomisation to maximise the statistical power of the variable selection procedure, measured through the true positive rate. Through extensive simulations, we show how our proposed strategy allows us to combine the benefits of the two techniques. The Mirror Statistic is a flexible method to control FDR, which only requires mild model assumptions, but requires two sets of independent regression coefficient estimates, usually obtained after splitting the original dataset. Outcome randomisation is an alternative to data splitting that allows to generate two independent outcomes, which can then be used to estimate the coefficients that go into the construction of the Mirror Statistic. The combination of these two approaches provides increased testing power in a number of scenarios, such as highly correlated covariates and high percentages of active variables. Moreover, it is scalable to very high-dimensional problems, since the algorithm has a low memory footprint and only requires a single run on the full dataset, as opposed to iterative alternatives such as multiple data splitting.
SMMR High-Dimensional Diagnostics & Prediction Survival Analysis · 发表于 2025-03-31
Fatih Kızılaslan, David Michael Swanson, Valeria Vitelli
摘要
A novel mixture cure frailty model is introduced for handling censored survival data. Mixture cure models are preferable when the existence of a cured fraction among patients can be assumed. However, such models are heavily underexplored: frailty structures within cure models remain largely undeveloped, and furthermore, most existing methods do not work for high-dimensional datasets, when the number of predictors is significantly larger than the number of observations. In this study, we introduce a novel extension of the Weibull mixture cure model that incorporates a frailty component, employed to model an underlying latent population heterogeneity with respect to the outcome risk. Additionally, high-dimensional covariates are integrated into both the cure rate and survival part of the model, providing a comprehensive approach to employ the model in the context of high-dimensional omics data. We also perform variable selection via an adaptive elastic-net penalization, and propose a novel approach to inference using the expectation–maximization (EM) algorithm. Extensive simulation studies are conducted across various scenarios to demonstrate the performance of the model, and results indicate that our proposed method outperforms competitor models. We apply the novel approach to analyze RNAseq gene expression data from bulk breast cancer patients included in The Cancer Genome Atlas (TCGA) database. A set of prognostic biomarkers is then derived from selected genes, and subsequently validated via both functional enrichment analysis and comparison to the existing biological literature. Finally, a prognostic risk score index based on the identified biomarkers is proposed and validated by exploring the patients’ survival.
SMMR Survival Analysis Causal Inference Clinical Trials · 发表于 2025-03-25
Marcel Wolbers, Mar Vázquez Rabuñal, Ke Li, Kaspar Rufibach, Daniel Sabanés Bové
摘要
In randomized controlled trials, forest plots are frequently used to investigate the homogeneity of treatment effect estimates in pre-defined subgroups. However, the interpretation of subgroup-specific treatment effect estimates requires great care due to the smaller sample size of subgroups and the large number of investigated subgroups. Bayesian shrinkage methods have been proposed to address these issues, but they often focus on disjoint subgroups while subgroups displayed in forest plots are overlapping, i.e., each subject appears in multiple subgroups. In our proposed approach, we first build a flexible Cox model based on all available observations, including treatment-by-subgroup interaction terms for all subgroups of interest. We explore penalized partial likelihood estimation with lasso or ridge penalties for interaction terms, and Bayesian estimation with a regularized horseshoe prior. In a second step, the Cox model is marginalized to obtain treatment effect estimates for all subgroups. We illustrate these methods using data from a randomized clinical trial in follicular lymphoma and evaluate their properties in a simulation study. In all simulation scenarios, the overall mean-squared error is substantially smaller for penalized and shrinkage estimators compared to the standard subgroup-specific treatment effect estimator but leads to some bias for heterogeneous subgroups. A naive overall sample estimator also outperforms the standard subgroup-specific estimator in terms of the overall mean-squared error for all scenarios except for a scenario with substantial heterogeneity. We recommend that subgroup-specific estimators are routinely complemented by treatment effect estimators based on shrinkage methods. The proposed methods are implemented in the R package bonsaiforest .
Biometrics Causal Inference · 发表于 2025-03-24
Manzoor Khan, Jake Olivier
摘要
Regression to the mean is said to have occurred when subjects having relatively high or low measurements are remeasured closer to the population mean. This phenomenon can influence the conclusion about the effectiveness of a treatment in a pre-post study design. The mean difference of the pre- and post-variables, conditioned on the initial variable being above or below a cut-point, is the sum of regression to the mean and treatment effects. Expressions for regression to the mean are available for the bivariate normal distribution under restrictive assumptions, and for the bivariate Poisson and binomial distributions, more generally. This article derives expressions for regression to the mean for any bivariate distribution while making fewer restrictive assumptions than previous methods. Maximum likelihood estimators are derived, and the unbiasedness, consistency, and asymptotic normality of these estimators are shown for exponential families, where possible. Data on the cholesterol levels in men aged 35-39 are used for decomposing the conditional mean difference in cholesterol level on pre-post occasions into regression to the mean and treatment effects. In another example, data on diastolic blood pressure for 341 patients are used to demonstrate the fraction of change due to regression to the mean and the treatment effects, respectively.
SiM Causal Inference · 发表于 2025-03-24,收录于 2026-06-12
Issa J. Dahabreh, James M. Robins, Sebastien J‐P. A. Haneuse, Sarah E. Robertson, Jon A. Steingrimsson, Miguel A. Hernán
摘要
When individuals participating in a randomized trial differ with respect to the distribution of effect modifiers compared with the target population where the trial results will be used, treatment effect estimates from the trial may not directly apply to target population. Methods for extending—generalizing or transporting—causal inferences from the trial to the target population rely on conditional exchangeability assumptions between randomized and non‐randomized individuals. The validity of these assumptions is often uncertain or controversial and investigators need to examine how violation of the assumptions would impact study conclusions. We describe methods for global sensitivity analysis that directly parameterize violations of the assumptions in terms of potential (counterfactual) outcome distributions. Our approach does not require detailed knowledge about the distribution of specific unmeasured effect modifiers or their relationship with the observed variables. We illustrate the methods using data from a trial nested within a cohort of trial‐eligible individuals to compare coronary artery surgery plus medical therapy versus medical therapy alone for stable ischemic heart disease.
SMMR Causal Inference Survival Analysis Nonparametric & Semiparametric · 发表于 2025-03-24
Jih-Chang Yu, Yen-Tsung Huang
摘要
We propose a causal mediation approach to semi-competing risks under left truncation sampling by considering an intermediate event as a mediator and a terminal event as an outcome. We focus on the causal relationship from exposure to the terminal outcome in relation to the intermediate event. In particular, we study the direct effect, the effect of exposure on the terminal event that is not through the intermediate event, and the indirect effect—the effect of exposure on the terminal event that is mediated through the intermediate event. We propose nonparametric and semiparametric methods, both accounting for left truncation. The nonparametric estimator can be viewed as a model-free time-varying Nelson–Aalen estimator that is robust to model misspecification. The semiparametric estimator calculated with the Cox proportional hazards model enjoys flexibility in adjusting for potential confounders as covariates. The asymptotic properties for both estimators, including uniform consistency and weak convergence, were established using the martingale theorem and functional delta method. The finite sample performance of the proposed estimators was evaluated through extensive numerical studies that investigated the influences of left truncation, confounding, and sample size. The utility of the proposed methods was illustrated using a hepatitis study.
Biometrics Spatial & Time Series · 发表于 2025-03-22
Raquel Menezes
摘要
In “Spatial Sampling with R,” Dick Brus provides a much-needed exploration of spatial sampling methods, filling a significant gap in the literature. While there are numerous books on statistical analysis and modeling using R, few address the foundational step of data collection through proper sampling design—a crucial aspect, since flawed sampling often undermines the validity of spatial analyses. Too often, researchers focus solely on data analysis, overlooking the critical importance of how the data are gathered. This book emphasizes the value of carefully planned spatial sampling as an essential precursor to rigorous scientific inquiry. The book provides a comprehensive guide to spatial sampling, and is divided into 2 main parts that cover a range of techniques from classical to advanced. Part I focuses on probability sampling for estimating population parameters, such as a mean, total, or a proportion of a population. These include familiar designs such as simple and systematic random sampling, as well as more specialized approaches such as cluster sampling and probability proportional to size sampling. Brus offers not only theoretical overviews but also in-depth, practical instructions on implementing these designs with clear R code examples, making it easier for both theoretical and applied users. This first part lays a robust foundation for readers seeking to understand how to make reliable statistical inferences about population parameters through probability sampling. Brus explains these methods with clarity, offering direct application examples and R code for hands-on practice. *(abstract via OpenAlex)*
Biometrics Spatial & Time Series · 发表于 2025-03-22
John Hughes
摘要
Spatial Linear Models for Environmental Data comprises twelve chapters and a two-page appendix of relevant matrix theorems. The first chapter provides, in just fifteen pages, an excellent introduction to areal and point-level spatial data, the sampling scheme necessary for doing principled inference and/or prediction for such data, and spatial linear models. The authors’ presentation of the areal and point-level data types is especially clear owing to the use of a few well-chosen motivating (real) datasets, some point-level and some areal. Chapter 1 also offers effective plots. Although many of the plots employ the color blind-friendly viridis palette (Garnier et al., 2024), some plots do not. The authors might consider strengthening their next edition by presenting only color blind-friendly plots. Chapter 2 describes covariance structures for spatial linear models (SLM). This is arguably the most important chapter in the book since it is precisely (often complicated) spatial dependence that distinguishes spatial data from other kinds of data. The authors provide an excellent introduction here, conveying the richness of the subject without including too many details at this stage. If I were using this text as a guide for a course, I would likely present plots of several point-level covariance functions from Chapter 6 hand-in-hand with those of Chapter 2. *(abstract via OpenAlex)*