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
2020 年的期刊文章 · 第 12 页
共 1467 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2020-08-12 到 2020-08-01。
按期刊、按主题看 →
SiM Survival Analysis Machine Learning Nonparametric & Semiparametric · 发表于 2020-08-12
Pingye Zhang, Junshui Ma, Xinqun Chen, Yue Shentu
摘要
In randomized clinical trials with survival outcome, there has been an increasing interest in subgroup identification based on baseline genomic, proteomic markers, or clinical characteristics. Some of the existing methods identify subgroups that benefit substantially from the experimental treatment by directly modeling outcomes or treatment effect. When the goal is to find an optimal treatment for a given patient rather than finding the right patient for a given treatment, methods under the individualized treatment regime framework estimate an individualized treatment rule that would lead to the best expected clinical outcome as measured by a value function. Connecting the concept of value function to subgroup identification, we propose a nonparametric method that searches for subgroup membership scores by maximizing a value function that directly reflects the subgroup‐treatment interaction effect based on restricted mean survival time. A gradient tree boosting algorithm is proposed to search for the individual subgroup membership scores. We conduct simulation studies to evaluate the performance of the proposed method and an application to an AIDS clinical trial is performed for illustration.
JASA Longitudinal Data Causal Inference Optimization & Learning Theory · 发表于 2020-08-11
Xinyu Hu, Min Qian, Bin Cheng, Ying Kuen Cheung
摘要
Personalized policy represents a paradigm shift from one-decision-rule-for-all users to an individualized decision rule for each user. Developing personalized policy in mobile health applications imposes challenges. First, for lack of adherence, data from each user are limited. Second, unmeasured contextual factors can potentially impact on decision making. Aiming to optimize immediate rewards, we propose using a generalized linear mixed modeling framework where population features and individual features are modeled as fixed and random effects, respectively, and synthesized to form the personalized policy. The group lasso type penalty is imposed to avoid overfitting of individual deviations from the population model. We examine the conditions under which the proposed method work in the presence of time-varying endogenous covariates, and provide conditional optimality and marginal consistency results of the expected immediate outcome under the estimated policies. We apply our method to develop personalized push ("prompt") schedules in 294 app users, with the goal to maximize the prompt response rate given past app usage and other contextual factors. The proposed method compares favorably to existing estimation methods including using the R function "glmer" in a simulation study. *(abstract via PubMed)*
SiM Bayesian Methods · 发表于 2020-08-11
Gail E. Potter
摘要
The Fragility Index has been introduced as a complement to theP‐value to summarize the statistical strength of evidence for a trial's result. The Fragility Index (FI) is defined in trials with two equal treatment group sizes, with a dichotomous or time‐to‐event outcome, and is calculated as the minimum number of conversions from nonevent to event in the treatment group needed to shift theP‐value from Fisher's exact test over the .05 threshold. As the index lacks a well‐defined probability motivation, its interpretation is challenging for consumers. We clarify what the FI may be capturing by separately considering two scenarios: (a) what the FI is capturing mathematically when the probability model is correct and (b) how well the FI captures violations of probability model assumptions. By calculating the posterior probability of a treatment effect, we show that when the probability model is correct, the FI inappropriately penalizes small trials for using fewer events than larger trials to achieve the same significance level. The analysis shows that for experiments conducted without bias, the FI promotes an incorrect intuition of probability, which has not been noted elsewhere and must be dispelled. We illustrate shortcomings of the FI's ability to quantify departures from model assumptions and contextualize the FI concept within current debate around the null hypothesis significance testing paradigm. Altogether, the FI creates more confusion than it resolves and does not promote statistical thinking. We recommend against its use. Instead, sensitivity analyses are recommended to quantify and communicate robustness of trial results.
SiM Other · 发表于 2020-08-11
Alexandra G. Ellis, Rowan Iskandar, Christopher H. Schmid, John B. Wong, Thomas A. Trikalinos
摘要
An emulator is a fast‐to‐evaluate statistical approximation of a detailed mathematical model (simulator). When used in lieu of simulators, emulators can expedite tasks that require many repeated evaluations, such as sensitivity analyses, policy optimization, model calibration, and value‐of‐information analyses. Emulators are developed using the output of simulators at specific input values (design points). Developing an emulator that closely approximates the simulator can require many design points, which becomes computationally expensive. We describe a self‐terminating active learning algorithm to efficiently develop emulators tailored to a specific emulation task, and compare it with algorithms that optimize geometric criteria (random latin hypercube sampling and maximum projection designs) and other active learning algorithms (treed Gaussian Processes that optimize typical active learning criteria). We compared the algorithms' root mean square error (RMSE) and maximum absolute deviation from the simulator (MAX) for seven benchmark functions and in a prostate cancer screening model. In the empirical analyses, in simulators with greatly varying smoothness over the input domain, active learning algorithms resulted in emulators with smaller RMSE and MAX for the same number of design points. In all other cases, all algorithms performed comparably. The proposed algorithm attained satisfactory performance in all analyses, had smaller variability than the treed Gaussian Processes, and, on average, had similar or better performance as the treed Gaussian Processes in six out of seven benchmark functions and in the prostate cancer model.
SiM Survival Analysis · 发表于 2020-08-10
Roland A. Matsouaka, Folefac D. Atem
摘要
Summary: In a longitudinal study, measures of key variables might be incomplete or partially recorded due to drop‐out, loss to follow‐up, or early termination of the study occurring before the advent of the event of interest. In this paper, we focus primarily on the implementation of a regression model with a randomly censored predictor. We examine, particularly, the use of inverse probability weighting methods in a generalized linear model (GLM), when the predictor of interest is right‐censored, to adjust for censoring. To improve the performance of the complete‐case analysis and prevent selection bias, we consider three different weighting schemes: inverse censoring probability weights, Kaplan‐Meier weights, and Cox proportional hazards weights. We use Monte Carlo simulation studies to evaluate and compare the empirical properties of different weighting estimation methods. Finally, we apply these methods to the Framingham Heart Study data as an illustrative example to estimate the relationship between age of onset of a clinically diagnosed cardiovascular event and low‐density lipoprotein among cigarette smokers.
SiM Clinical Trials · 发表于 2020-08-10
Dateng Li, Song Zhang, Jing Cao
摘要
Cluster randomized designs are frequently employed in pragmatic clinical trials which test interventions in the full spectrum of everyday clinical settings in order to maximize applicability and generalizability. In this study, we propose to directly incorporate pragmatic features into power analysis for cluster randomized trials with count outcomes. The pragmatic features considered include arbitrary randomization ratio, overdispersion, random variability in cluster size, and unequal lengths of follow‐up over which the count outcome is measured. The proposed method is developed based on generalized estimating equation (GEE) and it is advantageous in that the sample size formula retains a closed form, facilitating its implementation in pragmatic trials. We theoretically explore the impact of various pragmatic features on sample size requirements. An efficient Jackknife algorithm is presented to address the problem of underestimated variance by the GEE sandwich estimator when the number of clusters is small. We assess the performance of the proposed sample size method through extensive simulation and an application example to a real clinical trial is presented.
SiM Clinical Trials · 发表于 2020-08-10
Junxiao Hu, Patrick J. Blatchford, Neil A. Goldenberg, John M. Kittelson
摘要
Although all clinical trials are designed and monitored using more than one endpoint, methods are needed to assure that decision criteria are chosen to reflect the clinically relevant tradeoffs that assure the trial's scientific integrity. This article presents a framework for the design and monitoring clinical trials in a bivariate outcome space. The framework uses a rectangular hyperbola to define a bivariate null curve that divides outcome space into regions of benefit and lack of benefit. The curve is shown to be a flexible mapping of bivariate space that allows a continuous tradeoff between the two endpoints in a manner that captures many previous bivariate designs. The curve is extended to a distance function in bivariate space that allows different decisions in each of the four quadrants that comprise bivariate space. The distance function forms a statistic (); the distribution of its estimate is derived and used as a basis for trial design and group sequential monitoring plans in bivariate space. A recursive form of the bivariate group sequential density is used to evaluate and control operating characteristics for the proposed design. The bivariate designs are shown to meet or exceed the usual standards for size and power. The proposed design is illustrated in the ongoing NHLBI‐sponsored Kids‐DOTT multinational randomized controlled trial comparing shortened versus conventional anticoagulation for the treatment of venous thromboembolism in patients less than 21 years of age. The proposed methods are broadly applicable to a wide range of clinical settings and trial designs.
SMMR Longitudinal Data High-Dimensional Machine Learning · 发表于 2020-08-09
Louis Capitaine, Robin Genuer, Rodolphe Thiébaut
摘要
Random forests are one of the state-of-the-art supervised machine learning methods and achieve good performance in high-dimensional settings where p, the number of predictors, is much larger than n, the number of observations. Repeated measurements provide, in general, additional information, hence they are worth accounted especially when analyzing high-dimensional data. Tree-based methods have already been adapted to clustered and longitudinal data by using a semi-parametric mixed effects model, in which the non-parametric part is estimated using regression trees or random forests. We propose a general approach of random forests for high-dimensional longitudinal data. It includes a flexible stochastic model which allows the covariance structure to vary over time. Furthermore, we introduce a new method which takes intra-individual covariance into consideration to build random forests. Through simulation experiments, we then study the behavior of different estimation methods, especially in the context of high-dimensional data. Finally, the proposed method has been applied to an HIV vaccine trial including 17 HIV-infected patients with 10 repeated measurements of 20,000 gene transcripts and blood concentration of human immunodeficiency virus RNA. The approach selected 21 gene transcripts for which the association with HIV viral load was fully relevant and consistent with results observed during primary infection.
Biometrics Clinical Trials Survival Analysis Bayesian Methods · 发表于 2020-08-08
Yifei Zhang, Sha Cao, Chi Zhang, Ick Hoon Jin, Yong Zang
摘要
Early‐phase dose‐finding clinical trials are often subject to the issue of late‐onset outcomes. In phase I/II clinical trials, the issue becomes more intractable because toxicity and efficacy can be competing risk outcomes such that the occurrence of the first outcome will terminate the other one. In this paper, we propose a novel Bayesian adaptive phase I/II clinical trial design to address the issue of late‐onset competing risk outcomes. We use the continuation‐ratio model to characterize the trinomial response outcomes and the cause‐specific hazard rate method to model the competing‐risk survival outcomes. We treat the late‐onset outcomes as missing data and develop a Bayesian data augmentation method to impute the missing data from the observations. We also propose an adaptive dose‐finding algorithm to allocate patients and identify the optimal biological dose during the trial. Simulation studies show that the proposed design yields desirable operating characteristics.
SiM Survival Analysis Nonparametric & Semiparametric · 发表于 2020-08-07
Chenyang Zhang, Yuanshan Wu, Guosheng Yin
摘要
Restricted mean survival time (RMST) evaluates the mean event‐free survival time up to a prespecified time point. It has been used as an alternative measure of treatment effect owing to its model‐free structure and clinically meaningful interpretation of treatment benefit for right‐censored data. In clinical trials, another type of censoring called interval censoring may occur if subjects are examined at several discrete time points and the survival time falls into an interval rather than being exactly observed. The missingness of exact observations under interval‐censored cases makes the nonparametric measure of treatment effect more challenging. Employing the linear smoothing technique to overcome the ambiguity, we propose a new model‐free measure for the interval‐censored RMST. As an alternative to the commonly used log‐rank test, we further construct a hypothesis testing procedure to assess the survival difference between two groups. Simulation studies show that the bias of our proposed interval‐censored RMST estimator is negligible and the testing procedure delivers promising performance in detecting between‐group difference with regard to size and power under various configurations of survival curves. The proposed method is illustrated by reanalyzing two real datasets containing interval‐censored observations.
Biometrics Bayesian Methods Longitudinal Data Latent Variables & Mixtures · 发表于 2020-08-07
Carter Allen, Sara E. Benjamin‐Neelon, Brian Neelon
摘要
In studies of infant growth, an important research goal is to identify latent clusters of infants with delayed motor development—a risk factor for adverse outcomes later in life. However, there are numerous statistical challenges in modeling motor development: the data are typically skewed, exhibit intermittent missingness, and are correlated across repeated measurements over time. Using data from the Nurture study, a cohort of approximately 600 mother‐infant pairs, we develop a flexible Bayesian mixture model for the analysis of infant motor development. First, we model developmental trajectories using matrix skew‐normal distributions with cluster‐specific parameters to accommodate dependence and skewness in the data. Second, we model the cluster‐membership probabilities using a Pólya‐Gamma data‐augmentation scheme, which improves predictions of the cluster‐membership allocations. Lastly, we impute missing responses from conditional multivariate skew‐normal distributions. Bayesian inference is achieved through straightforward Gibbs sampling. Through simulation studies, we show that the proposed model yields improved inferences over models that ignore skewness or adopt conventional imputation methods. We applied the model to the Nurture data and identified two distinct developmental clusters, as well as detrimental effects of food insecurity on motor development. These findings can aid investigators in targeting interventions during this critical early‐life developmental window.
Biometrics Survival Analysis Optimization & Learning Theory · 发表于 2020-08-06
Rui Chen, Menggang Yu
摘要
A substantial rise in the number of cancer survivors has led to urgent management questions regarding effective posttreatment imaging‐based surveillance strategies for cancer recurrence. Current surveillance guidelines provided by a number of professional societies all warn against overly aggressive surveillance, especially for low‐risk patients, but all fail to provide more specific directions to accommodate underlying heterogeneity of cancer recurrence. Therefore it is imperative to develop data‐driven strategies that can tailor the surveillance schedules to recurrence risk in this era of stricter insurance regulations, provider shortages, and rising costs of health care. Due to a lack of statistical methods for optimizing surveillance scheduling in presence of competing risks, we propose a general approach that uses an intuitive loss function for optimization of early detection of recurrence before death. The proposed strategies can tailor to patient risks of recurrence, in terms of both intensity and amount of surveillance. Using general three‐state Markov models, our method is flexible and includes earlier works as special cases. We illustrate our method in both simulation studies and an application to breast cancer surveillance.
Biometrics Survival Analysis Networks & Graphs Missing Data & Measurement Error · 发表于 2020-08-06
Li‐Pang Chen, Grace Y. Yi
摘要
In survival data analysis, the Cox proportional hazards (PH) model is perhaps the most widely used model to feature the dependence of survival times on covariates. While many inference methods have been developed under such a model or its variants, those models are not adequate for handling data with complex structured covariates. High‐dimensional survival data often entail several features: (1) many covariates are inactive in explaining the survival information, (2) active covariates are associated in a network structure, and (3) some covariates are error‐contaminated. To hand such kinds of survival data, we propose graphical PH measurement error models and develop inferential procedures for the parameters of interest. Our proposed models significantly enlarge the scope of the usual Cox PH model and have great flexibility in characterizing survival data. Theoretical results are established to justify the proposed methods. Numerical studies are conducted to assess the performance of the proposed methods.
Biometrics Longitudinal Data Functional Data · 发表于 2020-08-05
Cody Carroll, Hans‐Georg Müller, Alois Kneip
摘要
Multivariate functional data are becoming ubiquitous with advances in modern technology and are substantially more complex than univariate functional data. We propose and study a novel model for multivariate functional data where the component processes are subject to mutual time warping. That is, the component processes exhibit a similar shape but are subject to systematic phase variation across their time domains. To address this previously unconsidered mode of warping, we propose new registration methodology that is based on a shift‐warping model. Our method differs from all existing registration methods for functional data in a fundamental way. Namely, instead of focusing on the traditional approach to warping, where one aims to recover individual‐specific registration, we focus on shift registration across the components of a multivariate functional data vector on a population‐wide level. Our proposed estimates for these shifts are identifiable, enjoy parametric rates of convergence, and often have intuitive physical interpretations, all in contrast to traditional curve‐specific registration approaches. We demonstrate the implementation and interpretation of the proposed method by applying our methodology to the Zürich Longitudinal Growth data and study its finite sample properties in simulations.
SMMR Bayesian Methods Networks & Graphs Meta-Analysis & Data Integration · 发表于 2020-08-05
Zhenxun Wang, Lifeng Lin, James S Hodges, Richard MacLehose, Haitao Chu
摘要
Network meta-analysis is a commonly used tool to combine direct and indirect evidence in systematic reviews of multiple treatments to improve estimation compared to traditional pairwise meta-analysis. Unlike the contrast-based network meta-analysis approach, which focuses on estimating relative effects such as odds ratios, the arm-based network meta-analysis approach can estimate absolute risks and other effects, which are arguably more informative in medicine and public health. However, the number of clinical studies involving each treatment is often small in a network meta-analysis, leading to unstable treatment-specific variance estimates in the arm-based network meta-analysis approach when using non- or weakly informative priors under an unequal variance assumption. Additional assumptions, such as equal (i.e. homogeneous) variances for all treatments, may be used to remedy this problem, but such assumptions may be inappropriately strong. This article introduces a variance shrinkage method for an arm-based network meta-analysis. Specifically, we assume different treatment variances share a common prior with unknown hyperparameters. This assumption is weaker than the homogeneous variance assumption and improves estimation by shrinking the variances in a data-dependent way. We illustrate the advantages of the variance shrinkage method by reanalyzing a network meta-analysis of organized inpatient care interventions for stroke. Finally, comprehensive simulations investigate the impact of different variance assumptions on statistical inference, and simulation results show that the variance shrinkage method provides better estimation for log odds ratios and absolute risks.
SiM Other · 发表于 2020-08-04
Chian Chen, Chin‐Fu Hsiao
摘要
A biosimilar is a biological product that is highly similar to an existing approved reference drug and has no clinically meaningful difference from it. Biosimilars are composed of or derived from living cells or organisms. Therefore, they are often sensitive to slight variations in the manufacturing process. Consequently, in demonstrating biosimilarity, it might be inappropriate to focus solely on the mean difference, or ratio of means, while ignoring the variabilities associated with the test and reference products. It is important to account for the entire population of clinical outcomes. Thus, we propose using the concept of tolerance intervals and related hypothesis testing for assessing biosimilarity. Our approach has the advantage of considering entire populations associated with both groups. A real example is used to illustrate our proposed method, and our approach is more stringent than those that employ confidence intervals. This is specifically the case when the mean difference of two drugs is not sufficiently large, but the biosimilar has a higher variability than that in the reference drug.
Biometrika Survey & Sampling Missing Data & Measurement Error · 发表于 2020-08-04
D N Da Silva, C J Skinner
摘要
Summary: Paradata refers to survey variables which are not of direct interest themselves, but are related to the quality of data on survey variables which are of interest. We focus on a categorical paradata variable, which reflects the presence of measurement error in a variable of interest. We propose a quasi-score test of the hypothesis of no measurement error bias in the estimation of regression coefficients under models for paradata. We also propose a regression-based test, analogous to a simple test proposed by Fuller for instrumental variables. The methods developed can take account of a complex sampling design. In an application with data from the British Household Panel Survey, all tests provide clear evidence of measurement bias in the estimated coefficient of gross pay. In a simulation study, all tests have rejection rates close to the nominal level under the null hypothesis; the quasi-score tests display more power than the regression-based test. The size of the quasi-score test is shown to be robust to some forms of misspecification of the paradata model, both by a theoretical argument and in findings of the simulation study.
SiM Survival Analysis · 发表于 2020-08-04
Philippe Flandre, John O'Quigley
摘要
We read with interest the paper, "Improving testing and description of treatment effect in clinical trials with survival outcomes" by Yang [1] and related works published by Yang and Prentice [2, 3]. Alternatives to absence of treatment effect on survival are most commonly framed within the proportional hazards model structure. When broader alternatives of a non-proportional hazards nature are of interest it is known that the high efficiency of the log-rank test can be lost. One test that focuses on this issue is the test of Yang and Prentice [3], based on the short-term and long-term hazard ratio model. The authors argue that the test is able to detect departures from a null hypothesis of no effect against quite broad alternatives. We recall the model on which this test is based and the test itself. Theoretical work and simulations appear to show the test to work well. However, it is important to underline that the test can only be used when the placebo group is clearly defined. Failing to correctly identify which group is the placebo group can have serious consequences and lead to erroneous conclusions. The results may not be interpretable. In situations where we wish to compare Treatment A with Treatment B, neither of which can be considered to correspond unambiguously to the placebo group, our advice would be to avoid this test altogether. We give two examples before concluding that the validity of this test hinges crucially upon a clear and sharp definition of the placebo group. *(abstract via OpenAlex)*
SiM Missing Data & Measurement Error · 发表于 2020-08-04
Kaifeng Lu
摘要
The standard multiple imputation technique focuses on parameter estimation. In this study, we describe a method for conducting score tests following multiple imputation. As an important application, we use the Cochran‐Mantel‐Haenszel (CMH) test as a score test and compare the proposed multiple imputation method with a method based on the Wilson‐Hilferty transformation of the CMH statistic. We show that the proposed multiple imputation method preserves the nominal significance level for three types of alternative hypotheses, whereas that based on the Wilson‐Hilferty transformation inflates type I error for the “row means differ” and “general association” alternative hypotheses. Moreover, we find that this type I error inflation worsens as the amount of missing data increases.
SiM Missing Data & Measurement Error Causal Inference · 发表于 2020-08-04
Dylan Spicker, Michael P. Wallace
摘要
Summary: Precision medicine incorporates patient‐level covariates to tailor treatment decisions, seeking to improve outcomes. In longitudinal studies with time‐varying covariates and sequential treatment decisions, precision medicine can be formalized with dynamic treatment regimes (DTRs): sequences of covariate‐dependent treatment rules. To date, the precision medicine literature has not addressed a ubiquitous concern in health research—measurement error—where observed data deviate from the truth. We discuss the consequences of ignoring measurement error in the context of DTRs, focusing on challenges unique to precision medicine. We show—through simulation and theoretical results—that relatively simple measurement error correction techniques can lead to substantial improvements over uncorrected analyses, and apply these findings to the sequenced treatment alternatives to relieve depression study.
Biometrika High-Dimensional Bayesian Methods · 发表于 2020-08-04
Ioannis Kosmidis, David Firth
摘要
Summary: Penalization of the likelihood by Jeffreys’ invariant prior, or a positive power thereof, is shown to produce finite-valued maximum penalized likelihood estimates in a broad class of binomial generalized linear models. The class of models includes logistic regression, where the Jeffreys-prior penalty is known additionally to reduce the asymptotic bias of the maximum likelihood estimator, and models with other commonly used link functions, such as probit and log-log. Shrinkage towards equiprobability across observations, relative to the maximum likelihood estimator, is established theoretically and studied through illustrative examples. Some implications of finiteness and shrinkage for inference are discussed, particularly when inference is based on Wald-type procedures. A widely applicable procedure is developed for computation of maximum penalized likelihood estimates, by using repeated maximum likelihood fits with iteratively adjusted binomial responses and totals. These theoretical results and methods underpin the increasingly widespread use of reduced-bias and similarly penalized binomial regression models in many applied fields.
SiM Nonparametric & Semiparametric Latent Variables & Mixtures · 发表于 2020-08-04
Zichen Ma, Timothy E. Hanson, Yen‐Yi Ho
摘要
Multivariate count data are common in many disciplines. The variables in such data often exhibit complex positive or negative dependency structures. We propose three Bayesian approaches to modeling bivariate count data by simultaneously considering covariate‐dependent means and correlation. A direct approach utilizes a bivariate negative binomial probability mass function developed in Famoye (2010,Journal of Applied Statistics). The second approach fits bivariate count data indirectly using a bivariate Poisson‐gamma mixture model. The third approach is a bivariate Gaussian copula model. Based on the results from simulation analyses, the indirect and copula approaches perform better overall than the direct approach in terms of model fitting and identifying covariate‐dependent association. The proposed approaches are applied to two RNA‐sequencing data sets for studying breast cancer and melanoma (BRCA‐US and SKCM‐US), respectively, obtained through the International Cancer Genome Consortium.
Biometrics High-Dimensional · 发表于 2020-08-04
Fang Fang, Jiwei Zhao, S. Ejaz Ahmed, Annie Qu
摘要
This paper is motivated from an HIV‐1 drug resistance study where we encounter three analytical challenges: to analyze data with an informative subsample, to take into account the weak signals, and to detect important signals and also conduct statistical inference. We start with an initial estimation method, which adopts a penalized pairwise conditional likelihood approach for variable selection. This initial estimator incorporates the informative subsample issue. To accounting for the effect of weak signals, we use a key idea of partial ridge regression. We also propose a one‐step estimation method for each of the signal coefficients and then construct confidence intervals accordingly. We apply the proposed method to the Stanford HIV‐1 drug resistance study and compare the results with existing approaches. We also conduct comprehensive simulation studies to demonstrate the superior performance of our proposed method.
SiM Missing Data & Measurement Error Testing & Uncertainty · 发表于 2020-08-04
Lak N. Kotinkaduwa, Pankaj K. Choudhary
摘要
Method comparison studies are concerned with estimating relationship between two clinical measurement methods. The methods often exhibit a structural change in the relationship over the measurement range. Ignoring this change would lead to an inaccurate estimate of the relationship. Motivated by a study of two digoxin assays where such a change occurs, this article develops a statistical methodology for appropriately analyzing such studies. Specifically, it proposes a segmented extension of the classical measurement error model to allow a piecewise linear relationship between the methods. The changepoint at which the transition takes place is treated as an unknown parameter in the model. An expectation‐maximization‐type algorithm is developed to fit the model and appropriate extensions of the existing measures are proposed for segment‐specific evaluation of similarity and agreement. Bootstrapping and large‐sample theory of maximum likelihood estimators are employed to perform the relevant inferences. The proposed methodology is evaluated by simulation and is illustrated by analyzing the digoxin data.
JASA Causal Inference Nonparametric & Semiparametric · 发表于 2020-08-04
Yifan Cui, Eric Tchetgen Tchetgen
摘要
There is a fast-growing literature on estimating optimal treatment regimes based on randomized trials or observational studies under a key identifying condition of no unmeasured confounding. Because confounding by unmeasured factors cannot generally be ruled out with certainty in observational studies or randomized trials subject to noncompliance, we propose a general instrumental variable approach to learning optimal treatment regimes under endogeneity. Specifically, we establish identification of both value function E [ Y D ( L ) ] for a given regime D and optimal regimes arg max D E [ Y D ( L ) ] with the aid of a binary instrumental variable, when no unmeasured confounding fails to hold. We also construct novel multiply robust classification-based estimators. Furthermore, we propose to identify and estimate optimal treatment regimes among those who would comply to the assigned treatment under a monotonicity assumption. In this latter case, we establish the somewhat surprising result that complier optimal regimes can be consistently estimated without directly collecting compliance information and therefore without the complier average treatment effect itself being identified. Our approach is illustrated via extensive simulation studies and a data application on the effect of child rearing on labor participation. *(abstract via PubMed)*
Biometrics Survival Analysis · 发表于 2020-08-03
Di Shu, Jessica G. Young, Sengwee Toh, Rui Wang
摘要
Inverse probability weighted Cox models can be used to estimate marginal hazard ratios under different point treatments in observational studies. To obtain variance estimates, the robust sandwich variance estimator is often recommended to account for the induced correlation among weighted observations. However, this estimator does not incorporate the uncertainty in estimating the weights and tends to overestimate the variance, leading to inefficient inference. Here we propose a new variance estimator that combines the estimation procedures for the hazard ratio and weights using stacked estimating equations, with additional adjustments for the sum of terms that are not independently and identically distributed in a Cox partial likelihood score equation. We prove analytically that the robust sandwich variance estimator is conservative and establish the asymptotic equivalence between the proposed variance estimator and one obtained through linearization by Hajage et al. in 2018. In addition, we extend our proposed variance estimator to accommodate clustered data. We compare the finite sample performance of the proposed method with alternative methods through simulation studies. We illustrate these different variance methods in both independent and clustered data settings, using a bariatric surgery dataset and a multiple readmission dataset, respectively. To facilitate implementation of the proposed method, we have developed an package ipwCoxCSV.
SMMR Missing Data & Measurement Error High-Dimensional Nonparametric & Semiparametric · 发表于 2020-08-03
Yongxin Bai, Maozai Tian, Man-Lai Tang, Wing-Yan Lee
摘要
In this paper, we consider variable selection for ultra-high dimensional quantile regression model with missing data and measurement errors in covariates. Specifically, we correct the bias in the loss function caused by measurement error by applying the orthogonal quantile regression approach and remove the bias caused by missing data using the inverse probability weighting. A nonconvex Atan penalized estimation method is proposed for simultaneous variable selection and estimation. With the proper choice of the regularization parameter and under some relaxed conditions, we show that the proposed estimate enjoys the oracle properties. The choice of smoothing parameters is also discussed. The performance of the proposed variable selection procedure is assessed by Monte Carlo simulation studies. We further demonstrate the proposed procedure with a breast cancer data set.
Biometrics Other · 发表于 2020-08-03
Brian L. Egleston, Tian Bai, Richard J. Bleicher, Stanford J. Taylor, Michael H. Lutz, Slobodan Vucetic
摘要
The pointwise mutual information statistic (PMI), which measures how often two words occur together in a document corpus, is a cornerstone of recently proposed popular natural language processing algorithms such as word2vec. PMI and word2vec reveal semantic relationships between words and can be helpful in a range of applications such as document indexing, topic analysis, or document categorization. We use probability theory to demonstrate the relationship between PMI and word2vec. We use the theoretical results to demonstrate how the PMI can be modeled and estimated in a simple and straight forward manner. We further describe how one can obtain standard error estimates that account for within‐patient clustering that arises from patterns of repeated words within a patient's health record due to a unique health history. We then demonstrate the usefulness of PMI on the problem of predictive identification of disease from free text notes of electronic health records. Specifically, we use our methods to distinguish those with and without type 2 diabetes mellitus in electronic health record free text data using over 400 000 clinical notes from an academic medical center.
JASA Causal Inference · 发表于 2020-08-03
Nathan Kallus
摘要
Nathan Kallus* School of Operations Research and Information Engineering and Cornell Tech, Cornell University, New York, NYColor versions of one or more of the figures in the article can be found online at www.tandfonline.com/r/JASA.Supplementary materials for this article are available online. Please go to www.tandfonline.com/r/JASA.CONTACT Nathan Kallus kallus@cornell.edu School of Operations Research and Information Engineering and Cornell Tech, Cornell University, 2 West Loop Rd, New York, NY 10044.AbstractPolicy learning can be used to extract individualized treatment regimes from observational data in healthcare, civics, e-commerce, and beyond. One big hurdle to policy learning is a commonplace lack of overlap in the data for different actions, which can lead to unwieldy policy evaluation and poorly performing learned policies. We study a solution to this problem based on retargeting, that is, changing the population on which policies are optimized. We first argue that at the population level, retargeting may induce little to no bias. We then characterize the optimal reference policy and retargeting weights in both binary-action and multi-action settings. We do this in terms of the asymptotic efficient estimation variance of the new learning objective. We further consider weights that additionally control for potential bias due to retargeting. Extensive empirical results in a simulation study and a case study of personalized job counseling demonstrate that retargeting is a fairly easy way to significantly improve any policy learning procedure applied to observational data. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
AoS Networks & Graphs Testing & Uncertainty High-Dimensional · 发表于 2020-08-01
Debarghya Ghoshdastidar, Maurilio Gutzeit, Alexandra Carpentier, Ulrike von Luxburg
摘要
The study of networks leads to a wide range of high-dimensional inference problems. In many practical applications, one needs to draw inference from one or few large sparse networks. The present paper studies hypothesis testing of graphs in this high-dimensional regime, where the goal is to test between two populations of inhomogeneous random graphs defined on the same set of $n$ vertices. The size of each population $m$ is much smaller than $n$, and can even be a constant as small as 1. The critical question in this context is whether the problem is solvable for small $m$. We answer this question from a minimax testing perspective. Let $P$, $Q$ be the population adjacencies of two sparse inhomogeneous random graph models, and $d$ be a suitably defined distance function. Given a population of $m$ graphs from each model, we derive minimax separation rates for the problem of testing $P=Q$ against $d(P,Q)>\rho $. We observe that if $m$ is small, then the minimax separation is too large for some popular choices of $d$, including total variation distance between corresponding distributions. This implies that some models that are widely separated in $d$ cannot be distinguished for small $m$, and hence, the testing problem is generally not solvable in these cases. We also show that if $m>1$, then the minimax separation is relatively small if $d$ is the Frobenius norm or operator norm distance between $P$ and $Q$. For $m=1$, only the latter distance provides small minimax separation. Thus, for these distances, the problem is solvable for small $m$. We also present near-optimal two-sample tests in both cases, where tests are adaptive with respect to sparsity level of the graphs.
AoS Causal Inference · 发表于 2020-08-01
Edward H. Kennedy, Sivaraman Balakrishnan, Max G’Sell
摘要
It is well known that, without restricting treatment effect heterogeneity, instrumental variable (IV) methods only identify “local” effects among compliers, that is, those subjects who take treatment only when encouraged by the IV. Local effects are controversial since they seem to only apply to an unidentified subgroup; this has led many to denounce these effects as having little policy relevance. However, we show that such pessimism is not always warranted: it can be possible to accurately predict who compliers are, and obtain tight bounds on more generalizable effects in identifiable subgroups. We propose methods for doing so and study estimation error and asymptotic properties, showing that these tasks can sometimes be accomplished even with very weak IVs. We go on to introduce a new measure of IV quality called “sharpness,” which reflects the variation in compliance explained by covariates, and captures how well one can identify compliers and obtain tight bounds on identifiable subgroup effects. We develop an estimator of sharpness and show that it is asymptotically efficient under weak conditions. Finally, we explore finite-sample properties via simulation, and apply the methods to study canvassing effects on voter turnout. We propose that sharpness should be presented alongside strength to assess IV quality.
AoS Bayesian Methods Optimization & Learning Theory · 发表于 2020-08-01
Veronika Ročková, Stéphanie van der Pas
摘要
Since their inception in the 1980s, regression trees have been one of the more widely used nonparametric prediction methods. Tree-structured methods yield a histogram reconstruction of the regression surface, where the bins correspond to terminal nodes of recursive partitioning. Trees are powerful, yet susceptible to overfitting. Strategies against overfitting have traditionally relied on pruning greedily grown trees. The Bayesian framework offers an alternative remedy against overfitting through priors. Roughly speaking, a good prior charges smaller trees where overfitting does not occur. While the consistency of random histograms, trees and their ensembles has been studied quite extensively, the theoretical understanding of the Bayesian counterparts has been missing. In this paper, we take a step toward understanding why/when do Bayesian trees and forests not overfit. To address this question, we study the speed at which the posterior concentrates around the true smooth regression function. We propose a spike-and-tree variant of the popular Bayesian CART prior and establish new theoretical results showing that regression trees (and forests) (a) are capable of recovering smooth regression surfaces (with smoothness not exceeding one), achieving optimal rates up to a log factor, (b) can adapt to the unknown level of smoothness and (c) can perform effective dimension reduction when $p>n$. These results provide a piece of missing theoretical evidence explaining why Bayesian trees (and additive variants thereof?) have worked so well in practice.
AoS Latent Variables & Mixtures · 发表于 2020-08-01
Yuqi Gu, Gongjun Xu
摘要
Latent class models have wide applications in social and biological sciences. In many applications, prespecified restrictions are imposed on the parameter space of latent class models, through a design matrix, to reflect practitioners’ assumptions about how the observed responses depend on subjects’ latent traits. Though widely used in various fields, such restricted latent class models suffer from nonidentifiability due to their discreteness nature and complex structure of restrictions. This work addresses the fundamental identifiability issue of restricted latent class models by developing a general framework for strict and partial identifiability of the model parameters. Under correct model specification, the developed identifiability conditions only depend on the design matrix and are easily checkable, which provide useful practical guidelines for designing statistically valid diagnostic tests. Furthermore, the new theoretical framework is applied to establish, for the first time, identifiability of several designs from cognitive diagnosis applications.
AoS Other · 发表于 2020-08-01
Yihong Wu, Pengkun Yang
摘要
The method of moments (Philos. Trans. R. Soc. Lond. Ser. A 185 (1894) 71–110) is one of the most widely used methods in statistics for parameter estimation, by means of solving the system of equations that match the population and estimated moments. However, in practice and especially for the important case of mixture models, one frequently needs to contend with the difficulties of non-existence or nonuniqueness of statistically meaningful solutions, as well as the high computational cost of solving large polynomial systems. Moreover, theoretical analyses of the method of moments are mainly confined to asymptotic normality style of results established under strong assumptions. This paper considers estimating a $k$-component Gaussian location mixture with a common (possibly unknown) variance parameter. To overcome the aforementioned theoretic and algorithmic hurdles, a crucial step is to denoise the moment estimates by projecting to the truncated moment space (via semidefinite programming) before solving the method of moments equations. Not only does this regularization ensure existence and uniqueness of solutions, it also yields fast solvers by means of Gauss quadrature. Furthermore, by proving new moment comparison theorems in the Wasserstein distance via polynomial interpolation and majorization techniques, we establish the statistical guarantees and adaptive optimality of the proposed procedure, as well as oracle inequality in misspecified models. These results can also be viewed as provable algorithms for generalized method of moments (Econometrica 50 (1982) 1029–1054) which involves nonconvex optimization and lacks theoretical guarantees.
AoS Bayesian Methods · 发表于 2020-08-01
John E. Kolassa, Todd A. Kuffner
摘要
We consider a fundamental open problem in parametric Bayesian theory, namely the validity of the formal Edgeworth expansion of the posterior density. While the study of valid asymptotic expansions for posterior distributions constitutes a rich literature, the validity of the formal Edgeworth expansion has not been rigorously established. Several authors have claimed connections of various posterior expansions with the classical Edgeworth expansion, or have simply assumed its validity. Our main result settles this open problem. We also prove a lemma concerning the order of posterior cumulants which is of independent interest in Bayesian parametric theory. The most relevant literature is synthesized and compared to the newly-derived Edgeworth expansions. Numerical investigations illustrate that our expansion has the behavior expected of an Edgeworth expansion, and that it has better performance than the other existing expansion which was previously claimed to be of Edgeworth type.
AoS Machine Learning Networks & Graphs Nonparametric & Semiparametric · 发表于 2020-08-01
Johannes Schmidt-Hieber
摘要
Consider the multivariate nonparametric regression model. It is shown that estimators based on sparsely connected deep neural networks with ReLU activation function and properly chosen network architecture achieve the minimax rates of convergence (up to $\log n$-factors) under a general composition assumption on the regression function. The framework includes many well-studied structural constraints such as (generalized) additive models. While there is a lot of flexibility in the network architecture, the tuning parameter is the sparsity of the network. Specifically, we consider large networks with number of potential network parameters exceeding the sample size. The analysis gives some insights into why multilayer feedforward neural networks perform well in practice. Interestingly, for ReLU activation function the depth (number of layers) of the neural network architectures plays an important role, and our theory suggests that for nonparametric regression, scaling the network depth with the sample size is natural. It is also shown that under the composition assumption wavelet estimators can only achieve suboptimal rates.
AoS Testing & Uncertainty · 发表于 2020-08-01
Mayya Zhilova
摘要
We study accuracy of bootstrap procedures for estimation of quantiles of a smooth function of a sum of independent sub-Gaussian random vectors. We establish higher-order approximation bounds with error terms depending on a sample size and a dimension explicitly. These results lead to improvements of accuracy of a weighted bootstrap procedure for general log-likelihood ratio statistics. The key element of our proofs of the bootstrap accuracy is a multivariate higher-order Berry–Esseen inequality. We consider a problem of approximation of distributions of two sums of zero mean independent random vectors, such that summands with the same indices have equal moments up to at least the second order. The derived approximation bound is uniform on the sets of all Euclidean balls. The presented approach extends classical Berry–Esseen type inequalities to higher-order approximation bounds. The theoretical results are illustrated with numerical experiments.
AoS High-Dimensional · 发表于 2020-08-01
Ching-Kang Ing
摘要
We investigate the prediction capability of the orthogonal greedy algorithm (OGA) in high-dimensional regression models with dependent observations. The rates of convergence of the prediction error of OGA are obtained under a variety of sparsity conditions. To prevent OGA from overfitting, we introduce a high-dimensional Akaike’s information criterion (HDAIC) to determine the number of OGA iterations. A key contribution of this work is to show that OGA, used in conjunction with HDAIC, can achieve the optimal convergence rate without knowledge of how sparse the underlying high-dimensional model is.