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
2018 年的期刊文章 · 第 11 页
共 1294 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2018-08-09 到 2018-08-02。
按期刊、按主题看 →
SiM Latent Variables & Mixtures · 发表于 2018-08-09
Yang Li, Ruofan Bie, Sebastian J Teran Hidalgo, Yichen Qin, Mengyun Wu, Shuangge Ma
摘要
In the research on complex diseases, gene expression (GE) data have been extensively used for clustering samples. The clusters so generated can serve as the basis for disease subtype identification, risk stratification, and many other purposes. With the small sample sizes of genetic profiling studies and noisy nature of GE data, clustering analysis results are often unsatisfactory. In the most recent studies, a prominent trend is to conduct multidimensional profiling, which collects data on GEs and their regulators (copy number alterations, microRNAs, methylation, etc.) on the same subjects. With the regulation relationships, regulators contain important information on the properties of GEs. We develop a novel assisted clustering method, which effectively uses regulator information to improve clustering analysis using GE data. To account for the fact that not all GEs are informative, we propose a weighted strategy, where the weights are determined data‐dependently and can discriminate informative GEs from noises. The proposed method is built on the NCut technique and effectively realized using a simulated annealing algorithm. Simulations demonstrate that it can well outperform multiple direct competitors. In the analysis of TCGA cutaneous melanoma and lung adenocarcinoma data, biologically sensible findings different from the alternatives are made.
SiM Bayesian Methods Latent Variables & Mixtures Genetics & Omics · 发表于 2018-08-09
Xue Li, Pankaj Kumar Choudhary, Swati Biswas, Xinlei Wang
摘要
In genomic research, it is becoming increasingly popular to perform meta‐analysis, the practice of combining results from multiple studies that target a common essential biological problem. Rank aggregation, a robust meta‐analytic approach, consolidates such studies at the rank level. There exists extensive research on this topic, and various methods have been developed in the past. However, these methods have two major limitations when they are applied in the genomic context. First, they are mainly designed to work with full lists, whereas partial and/or top‐ranked lists prevail in genomic studies. Second, the component studies are often clustered, and the existing methods fail to utilize such information. To address the above concerns, a Bayesian latent variable approach, called BiG, is proposed to formally deal with partial and top‐ranked lists and incorporate the effect of clustering. Various reasonable prior specifications for variance parameters in hierarchical models are carefully studied and compared. Simulation results demonstrate the superior performance of BiG compared with other popular rank aggregation methods under various practical settings. A non–small‐cell lung cancer data example is analyzed for illustration.
SiM Clinical Trials Testing & Uncertainty · 发表于 2018-08-08
Dean Follmann
摘要
Endpoint selection in clinical trials involves a variety of considerations. One important consideration is the sample size required to power a future clinical trial. In this work, we define the sample size ratio, θ, as the ratio of sample sizes required to power a future trial. We consider in detail the setting of continuous endpoints where a Welch's t‐statistic is used to analyze the data. We develop an estimator that depends on the squared ratio of estimated standardized treatment effects, and the quadrant on the plane in which they fall. We evaluate bootstrap and profile likelihood methods for construction of confidence intervals. Generalizations to other endpoints and testing of nonnested models are discussed. The methods are applied to analyze two different assays that measure antibody abundance using data from an Ebola vaccine field trial.
JASA Nonparametric & Semiparametric · 发表于 2018-08-07
Oscar Hernan Madrid Padilla, Alex Athey, Alex Reinhart, James G. Scott
摘要
We propose a sequential nonparametric test for detecting a change in distribution, based on windowed Kolmogorov–Smirnov statistics. The approach is simple, robust, highly computationally efficient, easy to calibrate, and requires no parametric assumptions about the underlying null and alternative distributions. We show that both the false-alarm rate and the power of our procedure are amenable to rigorous analysis, and that the method outperforms existing sequential testing procedures in practice. We then apply the method to the problem of detecting radiological anomalies, using data collected from measurements of the background gamma-radiation spectrum on a large university campus. In this context, the proposed method leads to substantial improvements in time-to-detection for the kind of radiological anomalies of interest in law-enforcement and border-security applications.Supplementary materials for this article, including a standardized description of the materials available for reproducing the work, are available as an online supplement. *(abstract via OpenAlex)*
SiM Genetics & Omics Survey & Sampling · 发表于 2018-08-07
Thea Bjørnland, Anja Bye, Einar Ryeng, Ulrik Wisløff, Mette Langaas
摘要
We consider cross‐sectional genetic association studies (common and rare variants) where non‐genetic information is available or feasible to obtain for N individuals, but where it is infeasible to genotype all N individuals. We consider continuously measurable Gaussian traits (phenotypes). Genotyping n < N extreme phenotype individuals can yield better power to detect phenotype‐genotype associations, as compared to randomly selecting n individuals. We define a person as having an extreme phenotype if the observed phenotype is above a specified threshold or below a specified threshold. We consider a model where these thresholds can be tailored to each individual. The classical extreme sampling design is to set equal thresholds for all individuals. We introduce a design (z‐extreme sampling) where personalized thresholds are defined based on the residuals of a regression model including only non‐genetic (fully available) information. We derive score tests for the situation where only n extremes are analyzed (complete case analysis) and for the situation where the non‐genetic information on N − n non‐extremes is included in the analysis (all case analysis). For the classical design, all case analysis is generally more powerful than complete case analysis. For the z‐extreme sample, we show that all case and complete case tests are equally powerful. Simulations and data analysis also show that z‐extreme sampling is at least as powerful as the classical extreme sampling design and the classical design is shown to be at times less powerful than random sampling. The method of dichotomizing extreme phenotypes is also discussed.
JASA Latent Variables & Mixtures · 发表于 2018-08-07
Denis Rybin, Robert Lew, Michael J. Pencina, Maurizio Fava, Gheorghe Doros
摘要
In clinical trials, placebo response can affect the inference about efficacy of the studied treatment. It is important to have a robust way to classify trial subjects with respect to their response to placebo. Simple, criterion-based classification may lead to classification error and bias the inference. The uncertainty about placebo response characteristic has to be factored into the treatment effect estimation. We propose a novel approach that views the placebo response as a latent characteristic and the study sample as an unlabeled mixture of “placebo responders” and “placebo nonresponders.” The likelihood-based methodology is used to estimate the treatment effect corrected for placebo response as defined within sequential parallel comparison design. *(abstract via OpenAlex)*
JASA Causal Inference Nonparametric & Semiparametric Longitudinal Data · 发表于 2018-08-07
Edward H. Kennedy
摘要
Most work in causal inference considers deterministic interventions that set each unit’s treatment to some fixed value. However, under positivity violations these interventions can lead to nonidentification, inefficiency, and effects with little practical relevance. Further, corresponding effects in longitudinal studies are highly sensitive to the curse of dimensionality, resulting in widespread use of unrealistic parametric models. We propose a novel solution to these problems: incremental interventions that shift propensity score values rather than set treatments to fixed values. Incremental interventions have several crucial advantages. First, they avoid positivity assumptions entirely. Second, they require no parametric assumptions and yet still admit a simple characterization of longitudinal effects, independent of the number of timepoints. For example, they allow longitudinal effects to be visualized with a single curve instead of lists of coefficients. After characterizing incremental interventions and giving identifying conditions for corresponding effects, we also develop general efficiency theory, propose efficient nonparametric estimators that can attain fast convergence rates even when incorporating flexible machine learning, and propose a bootstrap-based confidence band and simultaneous test of no treatment effect. Finally, we explore finite-sample performance via simulation, and apply the methods to study time-varying sociological effects of incarceration on entry into marriage. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
JASA High-Dimensional · 发表于 2018-08-07
Simon N. Vandekar, Philip T. Reiss, Russell T. Shinohara
摘要
In the fields of neuroimaging and genetics, a key goal is testing the association of a single outcome with a very high-dimensional imaging or genetic variable. Often, summary measures of the high-dimensional variable are created to sequentially test and localize the association with the outcome. In some cases, the associations between the outcome and summary measures are significant, but subsequent tests used to localize differences are underpowered and do not identify regions associated with the outcome. Here, we propose a generalization of Rao's score test based on projecting the score statistic onto a linear subspace of a high-dimensional parameter space. The approach provides a way to localize signal in the high-dimensional space by projecting the scores to the subspace where the score test was performed. This allows for inference in the high-dimensional space to be performed on the same degrees of freedom as the score test, effectively reducing the number of comparisons. Simulation results demonstrate the test has competitive power relative to others commonly used. We illustrate the method by analyzing a subset of the Alzheimer's Disease Neuroimaging Initiative dataset. Results suggest cortical thinning of the frontal and temporal lobes may be a useful biological marker of Alzheimer's disease risk. *(abstract via PubMed)*
JASA Bayesian Methods Spatial & Time Series High-Dimensional · 发表于 2018-08-07
Satyajit Ghosh, Kshitij Khare, George Michailidis
摘要
Vector autoregressive (VAR) models aim to capture linear temporal interdependencies amongst multiple time series. They have been widely used in macroeconomics and financial econometrics and more recently have found novel applications in functional genomics and neuroscience. These applications have also accentuated the need to investigate the behavior of the VAR model in a high-dimensional regime, which provides novel insights into the role of temporal dependence for regularized estimates of the model's parameters. However, hardly anything is known regarding properties of the posterior distribution for Bayesian VAR models in such regimes. In this work, we consider a VAR model with two prior choices for the autoregressive coefficient matrix: a non-hierarchical matrix-normal prior and a hierarchical prior, which corresponds to an arbitrary scale mixture of normals. We establish posterior consistency for both these priors under standard regularity assumptions, when the dimension p of the VAR model grows with the sample size n (but still remains smaller than n). A special case corresponds to a shrinkage prior that introduces (group) sparsity in the columns of the model coefficient matrices. The performance of the model estimates are illustrated on synthetic and real macroeconomic data sets. *(abstract via PubMed)*
JASA Other · 发表于 2018-08-07
Xinbing Kong, Jiangyan Wang, Jinbao Xing, Chao Xu, Chao Ying
摘要
The distributions of the common and idiosyncratic components for an individual variable are important in forecasting and applications. However, they are not identified with low-dimensional observations. Using the recently developed theory for large dimensional approximate factor model for large panel data, the common and idiosyncratic components can be estimated consistently. Based on the estimated common and idiosyncratic components, we construct the empirical processes for estimation of the distribution functions of the common and idiosyncratic components. We prove that the two empirical processes are oracle efficient when T = o(p) where p and T are the dimension and sample size, respectively. This demonstrates that the factor and idiosyncratic empirical processes behave as well as the empirical processes pretending that the common and idiosyncratic components for an individual variable are directly observable. Based on this oracle property, we construct simultaneous confidence bands (SCBs) for the distributions of the common and idiosyncratic components. For the first-order consistency of the estimated distribution functions, T=o(p) suffices. Extensive simulation studies check that the estimated bands have good coverage frequencies. Our real data analysis shows that the common-component distribution has a structural change during the crisis in 2008, while the idiosyncratic-component distribution does not change much. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
SiM Clinical Trials · 发表于 2018-08-07
Yi‐Da Chiu, Franz Koenig, Martin Posch, Thomas Jaki
摘要
Population heterogeneity is frequently observed among patients' treatment responses in clinical trials because of various factors such as clinical background, environmental, and genetic factors. Different subpopulations defined by those baseline factors can lead to differences in the benefit or safety profile of a therapeutic intervention. Ignoring heterogeneity between subpopulations can substantially impact on medical practice. One approach to address heterogeneity necessitates designs and analysis of clinical trials with subpopulation selection. Several types of designs have been proposed for different circumstances. In this work, we discuss a class of designs that allow selection of a predefined subgroup. Using the selection based on the maximum test statistics as the worst‐case scenario, we then investigate the precision and accuracy of the maximum likelihood estimator at the end of the study via simulations. We find that the required sample size is chiefly determined by the subgroup prevalence and show in simulations that the maximum likelihood estimator for these designs can be substantially biased.
SMMR Causal Inference Longitudinal Data · 发表于 2018-08-07
Fei Wan
摘要
The analysis of covariance (ANCOVA) or repeated measures (RM) models are often used to compare the treatment effect between different arms in pre-post randomized studies. ANCOVA adjusts the baseline score as a covariate in regression models. RM treats both the baseline and post-randomization scores as outcome variables. We aim to establish the underlying connections between ANCOVA and a constrained RM (“cRM”). We start with the interrelated concepts in a pre-post randomized designs: homogeneous vs. heterogeneous study populations, the marginal vs. the conditional treatment effect, and homogeneity vs. heterogeneity of treatment effect. We then demonstrate the asymptotic equivalence between the ANCOVA and cRM estimators for the marginal treatment effect and discuss the conditions under which ANCOVA needs to include a baseline score by treatment interaction term. In particular, an ANCOVA interaction model with a mean centered baseline score can assess both the marginal treatment effect and the heterogeneity in the conditional treatment effect. However, the ordinary least squares (OLS)-based inference is not valid for unconditional inference because this interaction model typically has heteroskedastic errors, and ordinary least squares treats the sample mean of the baseline score as a known parameter. We propose a bootstrap and a heteroskedasticity consistent variance estimator for heteroskedastic ANCOVA. Our simulation studies demonstrate that the proposed methods provide valid inferences for testing both the marginal treatment effect and the heterogeneity of treatment effect using an ANCOVA interaction model. We used an acupuncture headache trial to elucidate the proposed approaches.
JASA Meta-Analysis & Data Integration Bayesian Methods Networks & Graphs · 发表于 2018-08-07
Qinshu Lian, James S. Hodges, Haitao Chu
摘要
In studies evaluating the accuracy of diagnostic tests, three designs are commonly used, crossover, randomized, and non-comparative. Existing methods for meta-analysis of diagnostic tests mainly consider the simple cases in which the reference test in all or none of the studies can be considered a gold standard test, and in which all studies use either a randomized or non-comparative design. The proliferation of diagnostic instruments and the diversity of study designs create a need for more general methods to combine studies that include or do not include a gold standard test and that use various designs. This paper extends the Bayesian hierarchical summary receiver operating characteristic model to network meta-analysis of diagnostic tests to simultaneously compare multiple tests within a missing data framework. The method accounts for correlations between multiple tests and for heterogeneity between studies. It also allows different studies to include different subsets of diagnostic tests and provides flexibility in the choice of summary statistics. The model is evaluated using simulations and illustrated using real data on tests for deep vein thrombosis, with sensitivity analyses. *(abstract via PubMed)*
JASA Survival Analysis Causal Inference · 发表于 2018-08-06
Edward H. Kennedy, Steve Harris, Luke J. Keele
摘要
Pretreatment selection or censoring (“selection on treatment”) can occur when two treatment levels are compared ignoring the third option of neither treatment, in “censoring by death” settings where treatment is only defined for those who survive long enough to receive it, or in general in studies where the treatment is only defined for a subset of the population. Unfortunately, the standard instrumental variable (IV) estimand is not defined in the presence of such selection, so we consider estimating a new survivor-complier causal effect. Although this effect is generally not identified under standard IV assumptions, it is possible to construct sharp bounds. We derive these bounds and give a corresponding data-driven sensitivity analysis, along with nonparametric yet efficient estimation methods. Importantly, our approach allows for high-dimensional confounding adjustment, and valid inference even after employing machine learning. Incorporating covariates can tighten bounds dramatically, especially when they are strong predictors of the selection process. We apply the methods in a UK cohort study of critical care patients to examine the mortality effects of prompt admission to the intensive care unit, using ICU bed availability as an instrument. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
JASA Spatial & Time Series · 发表于 2018-08-06
Ganggang Xu, Rasmus Waagepetersen, Yongtao Guan
摘要
We propose a novel stochastic quasi-likelihood estimation procedure for case-control point processes. Quasi-likelihood for point processes depends on a certain optimal weight function and for the new method the weight function is stochastic since it depends on the control point pattern. The new procedure also provides a computationally efficient implementation of quasi-likelihood for univariate point processes in which case a synthetic control point process is simulated by the user. Under mild conditions, the proposed approach yields consistent and asymptotically normal parameter estimators. We further show that the estimators are optimal in the sense that the associated Godambe information is maximal within a wide class of estimating functions for case-control point processes. The effectiveness of the proposed method is further illustrated using extensive simulation studies and two data examples. *(abstract via OpenAlex)*
JASA Bayesian Methods Latent Variables & Mixtures · 发表于 2018-08-06
Jeffrey W. Miller, David B. Dunson
摘要
The standard approach to Bayesian inference is based on the assumption that the distribution of the data belongs to the chosen model class. However, even a small violation of this assumption can have a large impact on the outcome of a Bayesian procedure. We introduce a novel approach to Bayesian inference that improves robustness to small departures from the model: rather than conditioning on the event that the observed data are generated by the model, one conditions on the event that the model generates data close to the observed data, in a distributional sense. When closeness is defined in terms of relative entropy, the resulting "coarsened" posterior can be approximated by simply tempering the likelihood-that is, by raising the likelihood to a fractional power-thus, inference can usually be implemented via standard algorithms, and one can even obtain analytical solutions when using conjugate priors. Some theoretical properties are derived, and we illustrate the approach with real and simulated data using mixture models and autoregressive models of unknown order. *(abstract via PubMed)*
Biometrics Other · 发表于 2018-08-06
Ming Zhou, Rachel S. McCrea, Eleni Matechou, Diana J. Cole, Richard A. Griffiths
摘要
Summary: Removal of protected species from sites scheduled for development is often a legal requirement in order to minimize the loss of biodiversity. The assumption of closure in the classic removal model will be violated if individuals become temporarily undetectable, a phenomenon commonly exhibited by reptiles and amphibians. Temporary emigration can be modeled using a multievent framework with a partial hidden process, where the underlying state process describes the movement pattern of animals between the survey area and an area outside of the study. We present a multievent removal model within a robust design framework which allows for individuals becoming temporarily unavailable for detection. We demonstrate how to investigate parameter redundancy in the model. Results suggest the use of the robust design and certain forms of constraints overcome issues of parameter redundancy. We show which combinations of parameters are estimable when the robust design reduces to a single secondary capture occasion within each primary sampling period. Additionally, we explore the benefit of the robust design on the precision of parameters using simulation. We demonstrate that the use of the robust design is highly recommended when sampling removal data. We apply our model to removal data of common lizards, Zootoca vivipara, and for this application precision of parameter estimates is further improved using an integrated model.
SiM Clinical Trials · 发表于 2018-08-06
Zhaoyang Teng, Liang Liang, Guohui Liu, Yi Liu
摘要
In recent years, there has been an increasing trend in conducting seamless phase 2/3 clinical trials for drug development in the pharmaceutical industry due to the visible advantages compared with traditional approaches for separate phase 2 and 3 development. Innovative study designs have been proposed for seamless phase 2/3 trials, which mainly focus on the traditional aspects of study design, such as how to control the Type I error rate, sample size re‐estimation after the phase 2 portion, and how to make treatment selection at the end of the phase 2 portion. However, there are still some questions unresolved such as how to determine the go/no‐go boundary and how to plan the phase 2 portion sample size. In this paper, we discuss how to determine the phase 2 portion and phase 3 portion sample sizes as well as the corresponding go/no‐go criteria for a seamless phase 2/3 oncology trial based on Probability of Success. In addition, we further expand the methodology to include designs with an interim look within the phase 2 portion to speed up the go/no‐go decision‐making process. In case of overwhelming efficacy, this design would further shorten the overall trial duration. In case of inefficacy, this design can stop the trial earlier without exposing more patients to inefficacious treatment.
Biometrika Networks & Graphs Causal Inference · 发表于 2018-08-06
Guillaume W Basse, Edoardo M Airoldi
摘要
In this paper we consider how to assign treatment in a randomized experiment in which the correlation among the outcomes is informed by a network available pre-intervention. Working within the potential outcome causal framework, we develop a class of models that posit such a correlation structure among the outcomes. We use these models to develop restricted randomization strategies for allocating treatment optimally, by minimizing the mean squared error of the estimated average treatment effect. Analytical decompositions of the mean squared error, due both to the model and to the randomization distribution, provide insights into aspects of the optimal designs. In particular, the analysis suggests new notions of balance based on specific network quantities, in addition to classical covariate balance. The resulting balanced optimal restricted randomization strategies are still design-unbiased when the model used to derive them does not hold. We illustrate how the proposed treatment allocation strategies improve on allocations that ignore the network structure. *(abstract via OpenAlex)*
JASA Testing & Uncertainty Meta-Analysis & Data Integration · 发表于 2018-08-06
Patrick Rubin-Delanchy, Nicholas A. Heard, Daniel J. Lawson
摘要
The mid-p-value is a proposed improvement on the ordinary p-value for the case where the test statistic is partially or completely discrete. In this case, the ordinary p-value is conservative, meaning that its null distribution is larger than a uniform distribution on the unit interval, in the usual stochastic order. The mid-p-value is not conservative. However, its null distribution is dominated by the uniform distribution in a different stochastic order, called the convex order. The property leads us to discover some new finite-sample and asymptotic bounds on functions of mid-p-values, which can be used to combine results from different hypothesis tests conservatively, yet more powerfully, using mid-p-values rather than p-values. Our methodology is demonstrated on real data from a cyber-security application. *(abstract via PubMed)*
JASA High-Dimensional · 发表于 2018-08-06
Yuanpei Cao, Wei Lin, Hongzhe Li
摘要
Yuanpei Caoa, Wei Linb & Hongzhe Lia*a Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PAb School of Mathematical Sciences and Center for Statistical Science, Peking University, Beijing, ChinaCONTACT Hongzhe Li hongzhe@upenn.edu Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104.Color 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.ABSTRACTHigh-dimensional compositional data arise naturally in many applications such as metagenomic data analysis. The observed data lie in a high-dimensional simplex, and conventional statistical methods often fail to produce sensible results due to the unit-sum constraint. In this article, we address the problem of covariance estimation for high-dimensional compositional data and introduce a composition-adjusted thresholding (COAT) method under the assumption that the basis covariance matrix is sparse. Our method is based on a decomposition relating the compositional covariance to the basis covariance, which is approximately identifiable as the dimensionality tends to infinity. The resulting procedure can be viewed as thresholding the sample centered log-ratio covariance matrix and hence is scalable for large covariance matrices. We rigorously characterize the identifiability of the covariance parameters, derive rates of convergence under the spectral norm, and provide theoretical guarantees on support recovery. Simulation studies demonstrate that the COAT estimator outperforms some existing optimization-based estimators. We apply the proposed method to the analysis of a microbiome dataset to understand the dependence structure among bacterial taxa in the human gut. *(abstract via OpenAlex)*
Biometrics Networks & Graphs · 发表于 2018-08-06
Jianyu Liu, Wei Sun, Yufeng Liu
摘要
Summary: The directed acyclic graph (DAG) is a powerful tool to model the interactions of high-dimensional variables. While estimating edge directions in a DAG often requires interventional data, one can estimate the skeleton of a DAG (i.e., an undirected graph formed by removing the direction of each edge in a DAG) using observational data. In real data analyses, the samples of the high-dimensional variables may be collected from a mixture of multiple populations. Each population has its own DAG while the DAGs across populations may have significant overlap. In this article, we propose a two-step approach to jointly estimate the DAG skeletons of multiple populations while the population origin of each sample may or may not be labeled. In particular, our method allows a probabilistic soft label for each sample, which can be easily computed and often leads to more accurate skeleton estimation than hard labels. Compared with separate estimation of skeletons for each population, our method is more accurate and robust to labeling errors. We study the estimation consistency for our method, and demonstrate its performance using simulation studies in different settings. Finally, we apply our method to analyze gene expression data from breast cancer patients of multiple cancer subtypes.
JASA High-Dimensional Testing & Uncertainty · 发表于 2018-08-06
Damian Brzyski, Alexej Gossmann, Weijie Su, Małgorzata Bogdan
摘要
Sorted L-One Penalized Estimation (SLOPE, Bogdan et al., 2013, 2015) is a relatively new convex optimization procedure which allows for adaptive selection of regressors under sparse high dimensional designs. Here we extend the idea of SLOPE to deal with the situation when one aims at selecting whole groups of explanatory variables instead of single regressors. Such groups can be formed by clustering strongly correlated predictors or groups of dummy variables corresponding to different levels of the same qualitative predictor. We formulate the respective convex optimization problem, gSLOPE (group SLOPE), and propose an efficient algorithm for its solution. We also define a notion of the group false discovery rate (gFDR) and provide a choice of the sequence of tuning parameters for gSLOPE so that gFDR is provably controlled at a prespecified level if the groups of variables are orthogonal to each other. Moreover, we prove that the resulting procedure adapts to unknown sparsity and is asymptotically minimax with respect to the estimation of the proportions of variance of the response variable explained by regressors from different groups. We also provide a method for the choice of the regularizing sequence when variables in different groups are not orthogonal but statistically independent and illustrate its good properties with computer simulations. Finally, we illustrate the advantages of gSLOPE in the context of Genome Wide Association Studies. R package grpSLOPE with an implementation of our method is available on CRAN. *(abstract via PubMed)*
JASA Bayesian Methods · 发表于 2018-08-06
Yixin Wang, David M. Blei
摘要
A key challenge for modern Bayesian statistics is how to perform scalable inference of posterior distributions. To address this challenge, variational Bayes (VB) methods have emerged as a popular alternative to the classical Markov chain Monte Carlo (MCMC) methods. VB methods tend to be faster while achieving comparable predictive performance. However, there are few theoretical results around VB. In this article, we establish frequentist consistency and asymptotic normality of VB methods. Specifically, we connect VB methods to point estimates based on variational approximations, called frequentist variational approximations, and we use the connection to prove a variational Bernstein–von Mises theorem. The theorem leverages the theoretical characterizations of frequentist variational approximations to understand asymptotic properties of VB. In summary, we prove that (1) the VB posterior converges to the Kullback–Leibler (KL) minimizer of a normal distribution, centered at the truth and (2) the corresponding variational expectation of the parameter is consistent and asymptotically normal. As applications of the theorem, we derive asymptotic properties of VB posteriors in Bayesian mixture models, Bayesian generalized linear mixed models, and Bayesian stochastic block models. We conduct a simulation study to illustrate these theoretical results. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
JASA Other · 发表于 2018-08-06
Ryan J. Tibshirani, Saharon Rosset
摘要
Nearly all estimators in statistical prediction come with an associated tuning parameter, in one way or another. Common practice, given data, is to choose the tuning parameter value that minimizes a constructed estimate of the prediction error of the estimator; we focus on Stein’s unbiased risk estimator, or SURE, which forms an unbiased estimate of the prediction error by augmenting the observed training error with an estimate of the degrees of freedom of the estimator. Parameter tuning via SURE minimization has been advocated by many authors, in a wide variety of problem settings, and in general, it is natural to ask: what is the prediction error of the SURE-tuned estimator? An obvious strategy would be simply use the apparent error estimate as reported by SURE, that is, the value of the SURE criterion at its minimum, to estimate the prediction error of the SURE-tuned estimator. But this is no longer unbiased; in fact, we would expect that the minimum of the SURE criterion is systematically biased downwards for the true prediction error. In this work, we define the excess optimism of the SURE-tuned estimator to be the amount of this downward bias in the SURE minimum. We argue that the following two properties motivate the study of excess optimism: (i) an unbiased estimate of excess optimism, added to the SURE criterion at its minimum, gives an unbiased estimate of the prediction error of the SURE-tuned estimator; (ii) excess optimism serves as an upper bound on the excess risk, that is, the difference between the risk of the SURE-tuned estimator and the oracle risk (where the oracle uses the best fixed tuning parameter choice). We study excess optimism in two common settings: shrinkage estimators and subset regression estimators. Our main results include a James–Stein-like property of the SURE-tuned shrinkage estimator, which is shown to dominate the MLE; and both upper and lower bounds on excess optimism for SURE-tuned subset regression. In the latter setting, when the collection of subsets is nested, our bounds are particularly tight, and reveal that in the case of no signal, the excess optimism is always in between 0 and 10 degrees of freedom, regardless of how many models are being selected from. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
JASA Testing & Uncertainty · 发表于 2018-08-06
Prosper Dovonon, Sílvia Gonçalves, Ulrich Hounyo, Nour Meddahi
摘要
The main contribution of this article is to propose a bootstrap test for jumps based on functions of realized volatility and bipower variation. Bootstrap intraday returns are randomly generated from a mean zero Gaussian distribution with a variance given by a local measure of integrated volatility (which we denote by {v^in}). We first discuss a set of high-level conditions on {v^in} such that any bootstrap test of this form has the correct asymptotic size and is alternative-consistent. We then provide a set of primitive conditions that justify the choice of a thresholding-based estimator for {v^in}. Our cumulant expansions show that the bootstrap is unable to mimic the higher-order bias of the test statistic. We propose a modification of the original bootstrap test which contains an appropriate bias correction term and for which second-order asymptotic refinements are obtained. *(abstract via OpenAlex)*
Biometrics Bayesian Methods High-Dimensional Nonparametric & Semiparametric · 发表于 2018-08-06
Matthew W. Wheeler
摘要
Summary: Many modern datasets are sampled with error from complex high-dimensional surfaces. Methods such as tensor product splines or Gaussian processes are effective and well suited for characterizing a surface in two or three dimensions, but they may suffer from difficulties when representing higher dimensional surfaces. Motivated by high throughput toxicity testing where observed dose-response curves are cross sections of a surface defined by a chemical's structural properties, a model is developed to characterize this surface to predict untested chemicals’ dose-responses. This manuscript proposes a novel approach that models the multidimensional surface as a sum of learned basis functions formed as the tensor product of lower dimensional functions, which are themselves representable by a basis expansion learned from the data. The model is described and a Gibbs sampling algorithm is proposed. The approach is investigated in a simulation study and through data taken from the US EPA's ToxCast high throughput toxicity testing platform.
SiM Survival Analysis Missing Data & Measurement Error · 发表于 2018-08-06
Christopher A. Gravel, Anup Dewanji, Patrick J. Farrell, Daniel Krewski
摘要
Patient electronic health records, viewed as continuous‐time right‐censored survival data, can be used to estimate adverse drug reaction risk. Temporal outcome misclassification may occur as a result of errors in follow‐up. These errors can be due to a failure to observe the incidence time of the adverse event of interest (due to misdiagnosis or nonreporting, etc) or an actual misdiagnosis of a competing adverse event. As the misclassifying event is often unobservable in the original data, we apply an internal validation sampling approach to produce consistent estimation in the presence of such errors. We introduce a univariate survival model and a cause‐specific hazards model in which misclassification may also manifest as a diagnosis of an alternate adverse health outcome other than that of interest. We develop a method of maximum likelihood estimation of the model parameters and establish consistency and asymptotic normality of the estimators using standard results. We also conduct simulation studies to numerically investigate the finite sample properties of these estimators and the impact of ignoring the misclassification error.
JASA Bayesian Methods High-Dimensional · 发表于 2018-08-06
Roger S. Zoh, Abhra Sarkar, Raymond J. Carroll, Bani K. Mallick
摘要
We develop a Bayes factor based testing procedure for comparing two population means in high dimensional settings. In 'large-p-small-n' settings, Bayes factors based on proper priors require eliciting a large and complex p×p covariance matrix, whereas Bayes factors based on Jeffrey's prior suffer the same impediment as the classical Hotelling T 2 test statistic as they involve inversion of ill-formed sample covariance matrices. To circumvent this limitation, we propose that the Bayes factor be based on lower dimensional random projections of the high dimensional data vectors. We choose the prior under the alternative to maximize the power of the test for a fixed threshold level, yielding a restricted most powerful Bayesian test (RMPBT). The final test statistic is based on the ensemble of Bayes factors corresponding to multiple replications of randomly projected data. We show that the test is unbiased and, under mild conditions, is also locally consistent. We demonstrate the efficacy of the approach through simulated and real data examples. *(abstract via PubMed)*
JASA Other · 发表于 2018-08-06
Rong Zhu, Alan T. K. Wan, Xinyu Zhang, Guohua Zou
摘要
In the last decade, significant theoretical advances have been made in the area of frequentist model averaging (FMA); however, the majority of this work has emphasized parametric model setups. This article considers FMA for the semiparametric varying-coefficient partially linear model (VCPLM), which has gained prominence to become an extensively used modeling tool in recent years. Within this context, we develop a Mallows-type criterion for assigning model weights and prove its asymptotic optimality. A simulation study and a real data analysis demonstrate that the FMA estimator that arises from this criterion is vastly preferred to information criterion score-based model selection and averaging estimators. Our analysis is complicated by the fact that the VCPLM is subject to uncertainty arising not only from the choice of covariates, but also whether the covariate should enter the parametric or nonparametric parts of the model. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
JASA Other · 发表于 2018-08-06
Wenliang Pan, Xueqin Wang, Weinan Xiao, Hongtu Zhu
摘要
Extracting important features from ultra-high dimensional data is one of the primary tasks in statistical learning, information theory, precision medicine and biological discovery. Many of the sure independent screening methods developed to meet these needs are suitable for special models under some assumptions. With the availability of more data types and possible models, a model-free generic screening procedure with fewer and less restrictive assumptions is desirable. In this paper, we propose a generic nonparametric sure independence screening procedure, called BCor-SIS, on the basis of a recently developed universal dependence measure: Ball correlation. We show that the proposed procedure has strong screening consistency even when the dimensionality is an exponential order of the sample size without imposing sub-exponential moment assumptions on the data. We investigate the flexibility of this procedure by considering three commonly encountered challenging settings in biological discovery or precision medicine: iterative BCor-SIS, interaction pursuit, and survival outcomes. We use simulation studies and real data analyses to illustrate the versatility and practicability of our BCor-SIS method. *(abstract via PubMed)*
JASA Survey & Sampling Diagnostics & Prediction · 发表于 2018-08-06
Éric Lesage, David Haziza, Xavier D’Haultfœuille
摘要
Response rates have been steadily declining over the last decades, making survey estimates vulnerable to nonresponse bias. To reduce the potential bias, two weighting approaches are commonly used in National Statistical Offices: the one-step and the two-step approaches. In this article, we focus on the one-step approach, whereby the design weights are modified in a single step with two simultaneous goals in mind: reduce the nonresponse bias and ensure the consistency between survey estimates and known population totals. In particular, we examine the properties of instrumental calibration, a special case of the one-step approach that has received a lot of attention in the literature in recent years. Despite the rich literature on the topic, there remain some important gaps that this article aims to fill. First, we give a set of sufficient conditions required for establishing the consistency of instrumental calibration estimators. Also, we show that the latter may suffer from a large bias when some of these conditions are violated. Results from a simulation study support our findings. Supplementary materials for this article are available online. *(abstract via OpenAlex)*
Biometrika Testing & Uncertainty Latent Variables & Mixtures High-Dimensional · 发表于 2018-08-04
M Sesia, C Sabatti, E J Candès
摘要
Modern scientific studies often require the identification of a subset of explanatory variables. Several statistical methods have been developed to automate this task, and the framework of knockoffs has been proposed as a general solution for variable selection under rigorous Type I error control, without relying on strong modelling assumptions. In this paper, we extend the methodology of knockoffs to problems where the distribution of the covariates can be described by a hidden Markov model. We develop an exact and efficient algorithm to sample knockoff variables in this setting and then argue that, combined with the existing selective framework, this provides a natural and powerful tool for inference in genome-wide association studies with guaranteed false discovery rate control. We apply our method to datasets on Crohn's disease and some continuous phenotypes. *(abstract via PubMed)*
Biometrics Missing Data & Measurement Error Survival Analysis · 发表于 2018-08-04
Aurélie Bertrand, Ingrid Van Keilegom, Catherine Legrand
摘要
Summary: Measurement error in the continuous covariates of a model generally yields bias in the estimators. It is a frequent problem in practice, and many correction procedures have been developed for different classes of models. However, in most cases, some information about the measurement error distribution is required. When neither validation nor auxiliary data (e.g., replicated measurements) are available, this specification turns out to be tricky. In this article, we develop a flexible likelihood-based procedure to estimate the variance of classical additive error of Gaussian distribution, without additional information, when the covariate has compact support. The performance of this estimator is investigated both in an asymptotic way and through finite sample simulations. The usefulness of the obtained estimator when using the simulation extrapolation (SIMEX) algorithm, a widely used correction method, is then analyzed in the Cox proportional hazards model through other simulations. Finally, the whole procedure is illustrated on real data.
SMMR Other · 发表于 2018-08-03
Marianne A Jonker, Johannes A Rijken, Frederik J Hes, Hein Putter, Erik F Hensen
摘要
Accurate assessment of the age-dependent disease risk conferred by germline variants in disease susceptibility genes is often hampered by the way the data are collected. Cohort-based data sets frequently contain an overrepresentation of patients (i.e. carriers of the gene variant of interest affected with the associated disease), and an underrepresentation of disease-free carriers. In order to overcome this problem, penetrance estimates can be based on family-based study designs, through the evaluation of index patients and their family members. This approach facilitates the identification of asymptomatic germline variant carriers. By adjusting for the way these family data are ascertained, an estimate for the penetrance of the pathogenic gene variant can be obtained. However, the family structure is often incomplete or missing. This complicates the estimation of the penetrance, because full adjustment of the likelihood is not possible. We present a conditional likelihood for the estimation of the penetrance of pathogenic gene variants, based on a cohort of multiple families comprising index patients, disease-free and affected non-index carriers, but with missing information on pedigree structure. The proposed estimator corrects for the ascertainment in a robust way and is shown to be more accurate than the frequently used Kaplan–Meier estimator of the penetrance function.
SMMR Testing & Uncertainty Genetics & Omics · 发表于 2018-08-03
James J Yang, Elisa M Trucco, Anne Buu
摘要
Permutation tests are very useful when parametric assumptions are violated or distributions of test statistics are mathematically intractable. The major advantage of permutation tests is that the procedure is so general that it is applicable to most test statistics. The computational expense is, however, impractical in high-dimensional settings such as genomewide association studies. This study provides a comprehensive review of existing methods that can compute very small p-values efficiently. A common issue with existing methods is that they can only be applied to a specific test statistic. To fill in the knowledge gap, we propose a hybrid method of the sequential Monte Carlo and the Edgeworth expansion approximation for a studentized statistic, which is applicable to a variety of test statistics. The simulation results show that the proposed method performs better than competing methods. Furthermore, applications of the proposed method are demonstrated by statistical analysis on the genomewide association studies data from the Study of Addiction: Genetics and Environment (SAGE).
SiM Diagnostics & Prediction Causal Inference · 发表于 2018-08-02
Matthew Sperrin, Glen P. Martin, Alexander Pate, Tjeerd Van Staa, Niels Peek, Iain Buchan
摘要
Clinical prediction models (CPMs) can inform decision making about treatment initiation, which requires predicted risks assuming no treatment is given. However, this is challenging since CPMs are usually derived using data sets where patients received treatment, often initiated postbaseline as “treatment drop‐ins.” This study proposes the use of marginal structural models (MSMs) to adjust for treatment drop‐in. We illustrate the use of MSMs in the CPM framework through simulation studies that represent randomized controlled trials and real‐world observational data and the example of statin initiation for cardiovascular disease prevention. The simulations include a binary treatment and a covariate, each recorded at two timepoints and having a prognostic effect on a binary outcome. The bias in predicted risk was examined in a model ignoring treatment, a model fitted on treatment‐naïve patients (at baseline), a model including baseline treatment, and the MSM. In all simulation scenarios, all models except the MSM underestimated the risk of outcome given absence of treatment. These results were supported in the statin initiation example, which showed that ignoring statin initiation postbaseline resulted in models that significantly underestimated the risk of a cardiovascular disease event occurring within 10 years. Consequently, CPMs that do not acknowledge treatment drop‐in can lead to underallocation of treatment. In conclusion, when developing CPMs to predict treatment‐naïve risk, researchers should consider using MSMs to adjust for treatment drop‐in, and also seek to exploit the ability of MSMs to allow estimation of individual treatment effects.
SiM Longitudinal Data · 发表于 2018-08-02
Amna Klich, René Ecochard, Fabien Subtil
摘要
Classifying patients into groups according to longitudinal series of measurements (ie, trajectory classification) has become frequent in clinical research. Most classification models suppose an equal intra‐group variance across groups. This assumption is sometimes inappropriate because measurements in diseased subjects are often more heterogeneous than in healthy ones. We developed a new classification model for trajectories that uses unequal intra‐group variance across groups and evaluated its impact on classification using simulations and a clinical study. The classification and typical trajectories were estimated using the classification Expectation Maximization (EM) algorithm to maximize the classification likelihood, the log‐likelihood being profiled during the Maximization (M) step of the algorithm. The simulations showed that assuming equal intra‐group variance resulted in a high misclassification rate (up to 50%) when the real intra‐group variances were different. This rate was greatly reduced by allowing intra‐group variances to be different. Similar classification was obtained when the real intra‐group variances were equal, except when the total sample size and the number of repeated measurements were small. In a randomized trial that compared the effect of low vs standard cyclosporine A dose on creatinine levels after cardiac transplantation, the classification model with unequal intra‐group variance led to more meaningful groups than with equal intra‐group variance and showed distinct benefits of low dose. In conclusion, we recommend the use of a classification model for trajectories that allows for unequal intra‐group variance across groups except when the number of repeated measurements and total sample size are small.
Biometrics Causal Inference Survival Analysis · 发表于 2018-08-02
Matthias Brueckner, Andrew Titman, Thomas Jaki
摘要
Summary: Instrumental variable methods allow unbiased estimation in the presence of unmeasured confounders when an appropriate instrumental variable is available. Two-stage least-squares and residual inclusion methods have recently been adapted to additive hazard models for censored survival data. The semi-parametric additive hazard model which can include time-independent and time-dependent covariate effects is particularly suited for the two-stage residual inclusion method, since it allows direct estimation of time-independent covariate effects without restricting the effect of the residual on the hazard. In this article, we prove asymptotic normality of two-stage residual inclusion estimators of regression coefficients in a semi-parametric additive hazard model with time-independent and time-dependent covariate effects. We consider the cases of continuous and binary exposure. Estimation of the conditional survival function given observed covariates is discussed and a resampling scheme is proposed to obtain simultaneous confidence bands. The new methods are compared to existing ones in a simulation study and are applied to a real data set. The proposed methods perform favorably especially in cases with exposure-dependent censoring.
SiM Longitudinal Data · 发表于 2018-08-02
Whitney P. Ford, Philip M. Westgate
摘要
Data arising from longitudinal studies are commonly analyzed with generalized estimating equations. Previous literature has shown that liberal inference may result from the use of the empirical sandwich covariance matrix estimator when the number of subjects is small. Therefore, two different approaches have been used to improve the validity of inference. First, many different small‐sample corrections to the empirical estimator have been offered in order to reduce bias in resulting standard error estimates. Second, critical values can be obtained from a t‐distribution or an F‐distribution with approximated degrees of freedom. Although limited studies on the comparison of these small‐sample corrections and degrees of freedom have been published, there is a need for a comprehensive study of currently existing methods in a wider range of scenarios. Therefore, in this manuscript, we conduct such a simulation study, finding two methods to attain nominal type I error rates more consistently than other methods in a variety of settings: First, a recently proposed method by Westgate and Burchett (2016, Statistics in Medicine 35, 3733‐3744) that specifies both a covariance estimator and degrees of freedom, and second, an average of two popular corrections developed by Mancl and DeRouen (2001, Biometrics 57, 126‐134) and Kauermann and Carroll (2001, Journal of the American Statistical Association 96, 1387‐1396) with degrees of freedom equaling the number of subjects minus the number of parameters in the marginal model.