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 3 页
共 1294 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2018-12-01 到 2018-11-19。
按期刊、按主题看 →
AoS Other · 发表于 2018-12-01
Liping Zhu, Yaowu Zhang, Kai Xu
摘要
In this article, we introduce the notion of interval quantile independence which generalizes the notions of statistical independence and quantile independence. We suggest an index to measure and test departure from interval quantile independence. The proposed index is invariant to monotone transformations, nonnegative and equals zero if and only if the interval quantile independence holds true. We suggest a moment estimate to implement the test. The resultant estimator is root-$n$-consistent if the index is positive and $n$-consistent otherwise, leading to a consistent test of interval quantile independence. The asymptotic distribution of the moment estimator is free of parent distribution, which facilitates to decide the critical values for tests of interval quantile independence. When our proposed index is used to perform feature screening for ultrahigh dimensional data, it has the desirable sure screening property.
AoS Bayesian Methods Networks & Graphs Latent Variables & Mixtures · 发表于 2018-12-01
Robin J. Evans
摘要
Bayesian network models with latent variables are widely used in statistics and machine learning. In this paper, we provide a complete algebraic characterization of these models when the observed variables are discrete and no assumption is made about the state-space of the latent variables. We show that it is algebraically equivalent to the so-called nested Markov model, meaning that the two are the same up to inequality constraints on the joint probabilities. In particular, these two models have the same dimension, differing only by inequality constraints for which there is no general description. The nested Markov model is therefore the closest possible description of the latent variable model that avoids consideration of inequalities. A consequence of this is that the constraint finding algorithm of Tian and Pearl [In Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence (2002) 519–527] is complete for finding equality constraints. Latent variable models suffer from difficulties of unidentifiable parameters and nonregular asymptotics; in contrast the nested Markov model is fully identifiable, represents a curved exponential family of known dimension, and can easily be fitted using an explicit parameterization.
AoS Other · 发表于 2018-12-01
Mikael Escobar-Bach, Yuri Goegebeur, Armelle Guillou
摘要
We consider the robust estimation of the Pickands dependence function in the random covariate framework. Our estimator is based on local estimation with the minimum density power divergence criterion. We provide the main asymptotic properties, in particular the convergence of the stochastic process, correctly normalized, towards a tight centered Gaussian process. The finite sample performance of our estimator is evaluated with a simulation study involving both uncontaminated and contaminated samples. The method is illustrated on a dataset of air pollution measurements.
AoS High-Dimensional · 发表于 2018-12-01
Zhidong Bai, Kwok Pui Choi, Yasunori Fujikoshi
摘要
In this paper, we derive the asymptotic joint distributions of the eigenvalues under the null case and the local alternative cases in the MANOVA model and multiple discriminant analysis when both the dimension and the sample size are large. Our results are obtained by random matrix theory (RMT) without assuming normality in the populations. It is worth pointing out that the null and nonnull distributions of the eigenvalues and invariant test statistics are asymptotically robust against departure from normality in high-dimensional situations. Similar properties are pointed out for the null distributions of the invariant tests in multivariate regression model. Some new formulas in RMT are also presented.
AoS High-Dimensional · 发表于 2018-12-01
Preetam Nandy, Alain Hauser, Marloes H. Maathuis
摘要
Main approaches for learning Bayesian networks can be classified as constraint-based, score-based or hybrid methods. Although high-dimensional consistency results are available for constraint-based methods like the PC algorithm, such results have not been proved for score-based or hybrid methods, and most of the hybrid methods have not even shown to be consistent in the classical setting where the number of variables remains fixed and the sample size tends to infinity. In this paper, we show that consistency of hybrid methods based on greedy equivalence search (GES) can be achieved in the classical setting with adaptive restrictions on the search space that depend on the current state of the algorithm. Moreover, we prove consistency of GES and adaptively restricted GES (ARGES) in several sparse high-dimensional settings. ARGES scales well to sparse graphs with thousands of variables and our simulation study indicates that both GES and ARGES generally outperform the PC algorithm.
AoS Other · 发表于 2018-12-01
David Gamarnik, Quan Li
摘要
We consider the problem of finding a $k\times k$ submatrix of an $n\times n$ matrix with i.i.d. standard Gaussian entries, which has a large average entry. It was shown in [Bhamidi, Dey and Nobel (2012)] using nonconstructive methods that the largest average value of a $k\times k$ submatrix is $2(1+o(1))\sqrt{\log n/k}$, with high probability (w.h.p.), when $k=O(\log n/\log\log n)$. In the same paper, evidence was provided that a natural greedy algorithm called the Largest Average Submatrix ($\mathcal{LAS}$) for a constant $k$ should produce a matrix with average entry at most $(1+o(1))\sqrt{2\log n/k}$, namely approximately $\sqrt{2}$ smaller than the global optimum, though no formal proof of this fact was provided. In this paper, we show that the average entry of the matrix produced by the $\mathcal{LAS}$ algorithm is indeed $(1+o(1))\sqrt{2\log n/k}$ w.h.p. when $k$ is constant and $n$ grows. Then, by drawing an analogy with the problem of finding cliques in random graphs, we propose a simple greedy algorithm which produces a $k\times k$ matrix with asymptotically the same average value $(1+o(1))\sqrt{2\log n/k}$ w.h.p., for $k=o(\log n)$. Since the greedy algorithm is the best known algorithm for finding cliques in random graphs, it is tempting to believe that beating the factor $\sqrt{2}$ performance gap suffered by both algorithms might be very challenging. Surprisingly, we construct a very simple algorithm which produces a $k\times k$ matrix with average value $(1+o_{k}(1)+o(1))(4/3)\sqrt{2\log n/k}$ for $k=o((\log n)^{1.5})$, that is, with the asymptotic factor $4/3$ when $k$ grows. To get an insight into the algorithmic hardness of this problem, and motivated by methods originating in the theory of spin glasses, we conduct the so-called expected overlap analysis of matrices with average value asymptotically $(1+o(1))\alpha\sqrt{2\log n/k}$ for a fixed value $\alpha\in[1,\sqrt{2}]$. The overlap corresponds to the number of common rows and the number of common columns for pairs of matrices achieving this value (see the paper for details). We discover numerically an intriguing phase transition at $\alpha^{*}\triangleq5\sqrt{2}/(3\sqrt{3})\approx1.3608\ldots\in[4/3,\sqrt{2}]$: when $\alpha\alpha^{*}$, appropriately defined. We conjecture that the OGP observed for $\alpha>\alpha^{*}$ also marks the onset of the algorithmic hardness—no polynomial time algorithm exists for finding matrices with average value at least $(1+o(1))\alpha\sqrt{2\log n/k}$, when $\alpha>\alpha^{*}$ and $k$ is a mildly growing function of $n$.
AoS Other · 发表于 2018-12-01
Janet Godolphin
摘要
Designs with blocks of size two have numerous applications. In experimental situations where observation loss is common, it is important for a design to be robust against breakdown. For designs with one treatment factor and a single blocking factor, with blocks of size two, conditions for connectivity and robustness are obtained using combinatorial arguments and results from graph theory. Lower bounds are given for the breakdown number in terms of design parameters. For designs with equal or near equal treatment replication, the concepts of treatment and block partitions, and of linking blocks, are used to obtain information on the number of blocks required to guarantee various levels of robustness. The results provide guidance for construction of designs with good robustness properties. Robustness conditions are also established for row column designs in which one of the blocking factors involves blocks of size two. Such designs are particularly relevant for microarray experiments, where the high risk of observation loss makes robustness important. Disconnectivity in row column designs can be classified as three types. Techniques are given to assess design robustness according to each type, leading to lower bounds for the breakdown number. Guidance is given for robust design construction. Cyclic designs and interwoven loop designs are shown to have good robustness properties.
AoS High-Dimensional Clinical Trials · 发表于 2018-12-01
Adel Javanmard, Andrea Montanari
摘要
Performing statistical inference in high-dimensional models is challenging because of the lack of precise information on the distribution of high-dimensional regularized estimators. Here, we consider linear regression in the high-dimensional regime $p>>n$ and the Lasso estimator: we would like to perform inference on the parameter vector $\theta^{*}\in\mathbb{R}^{p}$. Important progress has been achieved in computing confidence intervals and $p$-values for single coordinates $\theta^{*}_{i}$, $i\in\{1,\dots,p\}$. A key role in these new inferential methods is played by a certain debiased estimator $\widehat{\theta}^{\mathrm{d}}$. Earlier work establishes that, under suitable assumptions on the design matrix, the coordinates of $\widehat{\theta}^{\mathrm{d}}$ are asymptotically Gaussian provided the true parameters vector $\theta^{*}$ is $s_{0}$-sparse with $s_{0}=o(\sqrt{n}/\log p)$. The condition $s_{0}=o(\sqrt{n}/\log p)$ is considerably stronger than the one for consistent estimation, namely $s_{0}=o(n/\log p)$. In this paper, we consider Gaussian designs with known or unknown population covariance. When the covariance is known, we prove that the debiased estimator is asymptotically Gaussian under the nearly optimal condition $s_{0}=o(n/(\log p)^{2})$. The same conclusion holds if the population covariance is unknown but can be estimated sufficiently well. For intermediate regimes, we describe the trade-off between sparsity in the coefficients $\theta^{*}$, and sparsity in the inverse covariance of the design. We further discuss several applications of our results beyond high-dimensional inference. In particular, we propose a thresholded Lasso estimator that is minimax optimal up to a factor $1+o_{n}(1)$ for i.i.d. Gaussian designs.
AoS Bayesian Methods · 发表于 2018-12-01
Bai Jiang, Tung-Yu Wu, Yifan Jin, Wing H. Wong
摘要
The Contrastive Divergence (CD) algorithm has achieved notable success in training energy-based models including Restricted Boltzmann Machines and played a key role in the emergence of deep learning. The idea of this algorithm is to approximate the intractable term in the exact gradient of the log-likelihood function by using short Markov chain Monte Carlo (MCMC) runs. The approximate gradient is computationally-cheap but biased. Whether and why the CD algorithm provides an asymptotically consistent estimate are still open questions. This paper studies the asymptotic properties of the CD algorithm in canonical exponential families, which are special cases of the energy-based model. Suppose the CD algorithm runs $m$ MCMC transition steps at each iteration $t$ and iteratively generates a sequence of parameter estimates $\{\theta_{t}\}_{t\ge 0}$ given an i.i.d. data sample $\{X_{i}\}_{i=1}^{n}\sim p_{\theta_{\star }}$. Under conditions which are commonly obeyed by the CD algorithm in practice, we prove the existence of some bounded $m$ such that any limit point of the time average $\sum_{s=0}^{t-1}\theta_{s}/t$ as $t\to\infty $ is a consistent estimate for the true parameter $\theta_{\star }$. Our proof is based on the fact that $\{\theta_{t}\}_{t\ge 0}$ is a homogenous Markov chain conditional on the data sample $\{X_{i}\}_{i=1}^{n}$. This chain meets the Foster–Lyapunov drift criterion and converges to a random walk around the maximum likelihood estimate. The range of the random walk shrinks to zero at rate $\mathcal{O}(1/\sqrt[3]{{n}})$ as the sample size $n\to \infty $.
AoS Causal Inference · 发表于 2018-12-01
Dominik Rothenhäusler, Jan Ernest, Peter Bühlmann
摘要
We consider identifiability of partially linear additive structural equation models with Gaussian noise (PLSEMs) and estimation of distributionally equivalent models to a given PLSEM. Thereby, we also include robustness results for errors in the neighborhood of Gaussian distributions. Existing identifiability results in the framework of additive SEMs with Gaussian noise are limited to linear and nonlinear SEMs, which can be considered as special cases of PLSEMs with vanishing nonparametric or parametric part, respectively. We close the wide gap between these two special cases by providing a comprehensive theory of the identifiability of PLSEMs by means of (A) a graphical, (B) a transformational, (C) a functional and (D) a causal ordering characterization of PLSEMs that generate a given distribution $\mathbb{P}$. In particular, the characterizations (C) and (D) answer the fundamental question to which extent nonlinear functions in additive SEMs with Gaussian noise restrict the set of potential causal models, and hence influence the identifiability. On the basis of the transformational characterization (B) we provide a score-based estimation procedure that outputs the graphical representation (A) of the distribution equivalence class of a given PLSEM. We derive its (high-dimensional) consistency and demonstrate its performance on simulated datasets.
AoS Other · 发表于 2018-12-01
Xavier Pennec
摘要
This paper investigates the generalization of Principal Component Analysis (PCA) to Riemannian manifolds. We first propose a new and general type of family of subspaces in manifolds that we call barycentric subspaces. They are implicitly defined as the locus of points which are weighted means of $k+1$ reference points. As this definition relies on points and not on tangent vectors, it can also be extended to geodesic spaces which are not Riemannian. For instance, in stratified spaces, it naturally allows principal subspaces that span several strata, which is impossible in previous generalizations of PCA. We show that barycentric subspaces locally define a submanifold of dimension $k$ which generalizes geodesic subspaces. Second, we rephrase PCA in Euclidean spaces as an optimization on flags of linear subspaces (a hierarchy of properly embedded linear subspaces of increasing dimension). We show that the Euclidean PCA minimizes the Accumulated Unexplained Variances by all the subspaces of the flag (AUV). Barycentric subspaces are naturally nested, allowing the construction of hierarchically nested subspaces. Optimizing the AUV criterion to optimally approximate data points with flags of affine spans in Riemannian manifolds lead to a particularly appealing generalization of PCA on manifolds called Barycentric Subspace Analysis (BSA).
Biometrics Other · 发表于 2018-11-29
Belmiro P.M. Duarte, Weng Kee Wong
摘要
Lot Quality Assurance Sampling (LQAS) plans are widely used for health monitoring purposes. We propose a systematic approach to design multiple-objective LQAS plans that meet user-specified type 1 and 2 error rates and targets for selected diagnostic accuracy metrics. These metrics may include sensitivity, specificity, positive predictive value, and negative predictive value in high or low anticipated prevalence rate populations. We use Mixed Integer Nonlinear Programming (MINLP) tools to implement our design methodology. Our approach is flexible in that it can directly generate classic LQAS plans that control error rates only and find optimal LQAS plans that meet multiple objectives in terms of diagnostic metrics. We give examples, compare results with the classic LQAS and provide an application using a malaria outcome indicator survey in Mozambique.
Biometrics Causal Inference · 发表于 2018-11-29
Issa J. Dahabreh, Sarah E. Robertson, Eric J. Tchetgen, Elizabeth A. Stuart, Miguel A. Hernán
摘要
We consider methods for causal inference in randomized trials nested within cohorts of trial-eligible individuals, including those who are not randomized. We show how baseline covariate data from the entire cohort, and treatment and outcome data only from randomized individuals, can be used to identify potential (counterfactual) outcome means and average treatment effects in the target population of all eligible individuals. We review identifiability conditions, propose estimators, and assess the estimators' finite-sample performance in simulation studies. As an illustration, we apply the estimators in a trial nested within a cohort of trial-eligible individuals to compare coronary artery bypass grafting surgery plus medical therapy vs. medical therapy alone for chronic coronary artery disease.
SiM Longitudinal Data · 发表于 2018-11-28
Eric O. Ohuma, Douglas G. Altman, for the International Fetal and Newborn Growth Consortium for the 21 Century
摘要
Most studies aiming to construct reference or standard charts use a cross‐sectional design, collecting one measurement per participant. Reference or standard charts can also be constructed using a longitudinal design, collecting multiple measurements per participant. The choice of appropriate statistical methodology is important as inaccurate centiles resulting from inferior methods can lead to incorrect judgements about fetal or newborn size, resulting in suboptimal clinical care. Reference or standard centiles should ideally provide the best fit to the data, change smoothly with age (eg, gestational age), use as simple a statistical model as possible without compromising model fit, and allow the computation of Z‐scores from centiles to simplify assessment of individuals and enable comparison with different populations. Significance testing and goodness‐of‐fit statistics are usually used to discriminate between models. However, these methods tend not to be useful when examining large data sets as very small differences are statistically significant even if the models are indistinguishable on actual centile plots. Choosing the best model from amongst many is therefore not trivial. Model choice should not be based on statistical considerations (or tests) alone as sometimes the best model may not necessarily offer the best fit to the raw data across gestational age. In this paper, we describe the most commonly applied methodologies available for the construction of age‐specific reference or standard centiles for cross‐sectional and longitudinal data: Fractional polynomial regression, LMS, LMST, LMSP, and multilevel regression methods. For illustration, we used data from the INTERGROWTH‐21 st Project, ie, newborn weight (cross‐sectional) and fetal head circumference (longitudinal) data as examples.
SiM Networks & Graphs Bayesian Methods Meta-Analysis & Data Integration · 发表于 2018-11-28
Rebecca M. Turner, Clara P. Domínguez‐Islas, Dan Jackson, Kirsty M. Rhodes, Ian R. White
摘要
In a network meta‐analysis, between‐study heterogeneity variances are often very imprecisely estimated because data are sparse, so standard errors of treatment differences can be highly unstable. External evidence can provide informative prior distributions for heterogeneity and, hence, improve inferences. We explore approaches for specifying informative priors for multiple heterogeneity variances in a network meta‐analysis. First, we assume equal heterogeneity variances across all pairwise intervention comparisons (approach 1); incorporating an informative prior for the common variance is then straightforward. Models allowing unequal heterogeneity variances are more realistic; however, care must be taken to ensure implied variance‐covariance matrices remain valid. We consider three strategies for specifying informative priors for multiple unequal heterogeneity variances. Initially, we choose different informative priors according to intervention comparison type and assume heterogeneity to be proportional across comparison types and equal within comparison type (approach 2). Next, we allow all heterogeneity variances in the network to differ, while specifying a common informative prior for each. We explore two different approaches to this: placing priors on variances and correlations separately (approach 3) or using an informative inverse Wishart distribution (approach 4). Our methods are exemplified through application to two network metaanalyses. Appropriate informative priors are obtained from previously published evidence‐based distributions for heterogeneity. Relevant prior information on between‐study heterogeneity can be incorporated into network meta‐analyses, without needing to assume equal heterogeneity across treatment comparisons. The approaches proposed will be beneficial in sparse data sets and provide more appropriate intervals for treatment differences than those based on imprecise heterogeneity estimates.
SiM Spatial & Time Series · 发表于 2018-11-28
Charlotte M. Jones‐Todd, Peter Caie, Janine B. Illian, Ben C. Stevenson, Anne Savage, David J. Harrison, James L. Bown
摘要
Diagnosis and prognosis of cancer are informed by the architecture inherent in cancer patient tissue sections. This architecture is typically identified by pathologists, yet advances in computational image analysis facilitate quantitative assessment of this structure. In this article, we develop a spatial point process approach to describe patterns in cell distribution within tissue samples taken from colorectal cancer (CRC) patients. In particular, our approach is centered on the Palm intensity function. This leads to taking an approximate‐likelihood technique in fitting point processes models. We consider two Neyman‐Scott point processes and a void process, fitting these point process models to the CRC patient data. We find that the parameter estimates of these models may be used to quantify the spatial arrangement of cells. Importantly, we observe characteristic differences in the spatial arrangement of cells between patients who died from CRC and those alive at follow up.
SiM Bayesian Methods Nonparametric & Semiparametric Latent Variables & Mixtures · 发表于 2018-11-28
Kai Kang, Xinyuan Song, X. Joan Hu, Hongtu Zhu
摘要
This paper presents a Bayesian adaptive group least absolute shrinkage and selection operator method to conduct simultaneous model selection and estimation under semiparametric hidden Markov models. We specify the conditional regression model and the transition probability model in the hidden Markov model into additive nonparametric functions of covariates. A basis expansion is adopted to approximate the nonparametric functions. We introduce multivariate conditional Laplace priors to impose adaptive penalties on regression coefficients and different groups of basis expansions under the Bayesian framework. An efficient Markov chain Monte Carlo algorithm is then proposed to identify the nonexistent, constant, linear, and nonlinear forms of covariate effects in both conditional and transition models. The empirical performance of the proposed methodology is evaluated via simulation studies. We apply the proposed model to analyze a real data set that was collected from the Alzheimer's Disease Neuroimaging Initiative study. The analysis identifies important risk factors on cognitive decline and the transition from cognitive normal to Alzheimer's disease.
SiM Survival Analysis Latent Variables & Mixtures Bayesian Methods · 发表于 2018-11-28
Brian Neelon, Azza Shoaibi, Sara E. Benjamin‐Neelon
摘要
We develop a multivariate discrete failure time model for the analysis of infant motor development. We use the model to jointly evaluate the time (in months) to achievement of three well‐established motor milestones: sitting up, crawling, and walking. The model includes a subject‐specific latent factor that reflects underlying heterogeneity in the population and accounts for within‐subject dependence across the milestones. The factor loadings and covariate effects are allowed to vary flexibly across milestones, and the milestones are permitted to have unique at‐risk intervals corresponding to different developmental windows. We adopt a Bayesian inferential approach and develop a convenient data‐augmented Gibbs sampler for posterior computation. We conduct simulation studies to illustrate key features of the model and use the model to analyze data from the Nurture study, a birth cohort examining infant health and development during the first year of life.
SiM Survey & Sampling · 发表于 2018-11-28
Jihnhee Yu, Albert Vexler, Kabir Jalal
摘要
Publicly available national survey data are useful for the evidence‐based research to advance our understanding of important questions in the health and biomedical sciences. Appropriate variance estimation is a crucial step to evaluate the strength of evidence in the data analysis. In survey data analysis, the conventional linearization method for estimating the variance of a statistic of interest uses the variance estimator of the total based on linearized variables. We warn that this common practice may result in undesirable consequences such as susceptibility to data shift and severely inflated variance estimates, when unequal weights are incorporated into variance estimation. We propose to use the variance estimator of the mean (mean‐approach) instead of the variance estimator of the total (total‐approach). We show a superiority of the mean‐approach through analytical investigations. A real data example (the National Comorbidity Survey Replication) and simulation‐based studies strongly support our conclusion.
SiM Causal Inference Bayesian Methods Genetics & Omics · 发表于 2018-11-28
Chin Yang Shapland, John R. Thompson, Nuala A. Sheehan
摘要
Mendelian randomisation (MR) is a method for establishing causality between a risk factor and an outcome by using genetic variants as instrumental variables. In practice, the association between individual genetic variants and the risk factor is often weak, which may lead to a lack of precision in the MR and even biased MR estimates. Usually, the most significant variant within a genetic region is selected to represent the association with the risk factor, but there is no guarantee that this variant will be causal or that it will capture all of the genetic association within the region. It may be advantageous to use extra variants selected from the same region in the MR. The problem is to decide which variants to select. Rather than selecting a specific set of variants, we investigate the use of Bayesian model averaging (BMA) to average the MR over all possible combinations of genetic variants. Our simulations demonstrate that the BMA version of MR outperforms classical estimation with many dependent variants and performs much better than an MR based on variants selected by penalised regression. In further simulations, we investigate robustness to violations in the model assumptions and demonstrate sensitivity to the inclusion of invalid instruments. The method is illustrated by applying it to an MR of the effect of body mass index on blood pressure using SNPs in the FTO gene.
SMMR Latent Variables & Mixtures · 发表于 2018-11-26
Peng Ye, Wan Tang, Jiang He, Hua He
摘要
Count outcomes with excessive zeros are common in behavioral and social studies, and zero-inflated count models such as zero-inflated Poisson (ZIP) and zero-inflated Negative Binomial (ZINB) can be applied when such zero-inflated count data are used as response variable. However, when the zero-inflated count data are used as predictors, ignoring the difference of structural and random zeros can result in biased estimates. In this paper, a generalized estimating equation (GEE)-type mixture model is proposed to jointly model the response of interest and the zero-inflated count predictors. Simulation studies show that the proposed method performs well for practical settings and is more robust for model misspecification than the likelihood-based approach. A case study is also provided for illustration.
SMMR Latent Variables & Mixtures · 发表于 2018-11-25
Theodora S Brisimi, Tingting Xu, Taiyao Wang, Wuyang Dai, Ioannis Ch Paschalidis
摘要
Objective: To derive a predictive model to identify patients likely to be hospitalized during the following year due to complications attributed to Type II diabetes. Methods: A variety of supervised machine learning classification methods were tested and a new method that discovers hidden patient clusters in the positive class (hospitalized) was developed while, at the same time, sparse linear support vector machine classifiers were derived to separate positive samples from the negative ones (non-hospitalized). The convergence of the new method was established and theoretical guarantees were proved on how the classifiers it produces generalize to a test set not seen during training. Results: The methods were tested on a large set of patients from the Boston Medical Center – the largest safety net hospital in New England. It is found that our new joint clustering/classification method achieves an accuracy of 89% (measured in terms of area under the ROC Curve) and yields informative clusters which can help interpret the classification results, thus increasing the trust of physicians to the algorithmic output and providing some guidance towards preventive measures. While it is possible to increase accuracy to 92% with other methods, this comes with increased computational cost and lack of interpretability. The analysis shows that even a modest probability of preventive actions being effective (more than 19%) suffices to generate significant hospital care savings. Conclusions: Predictive models are proposed that can help avert hospitalizations, improve health outcomes and drastically reduce hospital expenditures. The scope for savings is significant as it has been estimated that in the USA alone, about $5.8 billion are spent each year on diabetes-related hospitalizations that could be prevented.
SMMR Causal Inference · 发表于 2018-11-25
Andrew M Ryan, Evangelos Kontopantelis, Ariel Linden, James F Burgess
摘要
Difference-in-differences (DID) analysis is used widely to estimate the causal effects of health policies and interventions. A critical assumption in DID is “parallel trends”: that pre-intervention trends in outcomes are the same between treated and comparison groups. To date, little guidance has been available to researchers who wish to use DID when the parallel trends assumption is violated. Using a Monte Carlo simulation experiment, we tested the performance of several estimators (standard DID; DID with propensity score matching; single-group interrupted time-series analysis; and multi-group interrupted time-series analysis) when the parallel trends assumption is violated. Using nationwide data from US hospitals (n = 3737) for seven data periods (four pre-interventions and three post-interventions), we used alternative estimators to evaluate the effect of a placebo intervention on common outcomes in health policy (clinical process quality and 30-day risk-standardized mortality for acute myocardial infarction, heart failure, and pneumonia). Estimator performance was assessed using mean-squared error and estimator coverage. We found that mean-squared error values were considerably lower for the DID estimator with matching than for the standard DID or interrupted time-series analysis models. The DID estimator with matching also had superior performance for estimator coverage. Our findings were robust across all outcomes evaluated.
SiM Survival Analysis · 发表于 2018-11-25
Richard Tawiah, Kelvin K.W. Yau, Geoffrey J. McLachlan, Suzanne K. Chambers, Shu‐Kay Ng
摘要
We present a multilevel frailty model for handling serial dependence and simultaneous heterogeneity in survival data with a multilevel structure attributed to clustering of subjects and the presence of multiple failure outcomes. One commonly observes such data, for example, in multi‐institutional, randomized placebo‐controlled trials in which patients suffer repeated episodes (eg, recurrent migraines) of the disease outcome being measured. The model extends the proportional hazards model by incorporating a random covariate and unobservable random institution effect to respectively account for treatment‐by‐institution interaction and institutional variation in the baseline risk. Moreover, a random effect term with correlation structure driven by a first‐order autoregressive process is attached to the model to facilitate estimation of between patient heterogeneity and serial dependence. By means of the generalized linear mixed model methodology, the random effects distribution is assumed normal and the residual maximum likelihood and the maximum likelihood methods are extended for estimation of model parameters. Simulation studies are carried out to evaluate the performance of the residual maximum likelihood and the maximum likelihood estimators and to assess the impact of misspecifying random effects distribution on the proposed inference. We demonstrate the practical feasibility of the modeling methodology by analyzing real data from a double‐blind randomized multi‐institutional clinical trial, designed to examine the effect of rhDNase on the occurrence of respiratory exacerbations among patients with cystic fibrosis.
SMMR Genetics & Omics Econometrics & Economics Bayesian Methods · 发表于 2018-11-25
Viktor Jonsson, Tobias Österlund, Olle Nerman, Erik Kristiansson
摘要
Metagenomics enables the study of gene abundances in complex mixtures of microorganisms and has become a standard methodology for the analysis of the human microbiome. However, gene abundance data is inherently noisy and contains high levels of biological and technical variability as well as an excess of zeros due to non-detected genes. This makes the statistical analysis challenging. In this study, we present a new hierarchical Bayesian model for inference of metagenomic gene abundance data. The model uses a zero-inflated overdispersed Poisson distribution which is able to simultaneously capture the high gene-specific variability as well as zero observations in the data. By analysis of three comprehensive datasets, we show that zero-inflation is common in metagenomic data from the human gut and, if not correctly modelled, it can lead to substantial reductions in statistical power. We also show, by using resampled metagenomic data, that our model has, compared to other methods, a higher and more stable performance for detecting differentially abundant genes. We conclude that proper modelling of the gene-specific variability, including the excess of zeros, is necessary to accurately describe gene abundances in metagenomic data. The proposed model will thus pave the way for new biological insights into the structure of microbial communities.
SiM Clinical Trials Machine Learning · 发表于 2018-11-25
Zhiwei Zhang, Shujie Ma
摘要
Clinical trials are widely considered the gold standard for treatment evaluation, and they can be highly expensive in terms of time and money. The efficiency of clinical trials can be improved by incorporating information from baseline covariates that are related to clinical outcomes. This can be done by modifying an unadjusted treatment effect estimator with an augmentation term that involves a function of covariates. The optimal augmentation is well characterized in theory but must be estimated in practice. In this article, we investigate the use of machine learning methods to estimate the optimal augmentation. We consider and compare an indirect approach based on an estimated regression function and a direct approach that aims directly to minimize the asymptotic variance of the treatment effect estimator. Theoretical considerations and simulation results indicate that the direct approach is generally preferable over the indirect approach. The direct approach can be implemented using any existing prediction algorithm that can minimize a weighted sum of squared prediction errors. Many such prediction algorithms are available, and the super learning principle can be used to combine multiple algorithms into a super learner under the direct approach. The resulting direct super learner has a desirable oracle property, is easy to implement, and performs well in realistic settings. The proposed methodology is illustrated with real data from a stroke trial.
SMMR Bayesian Methods · 发表于 2018-11-25
Janet van Niekerk, Andriette Bekker, Mohammad Arashi
摘要
Real phenomena often leads to challenges in data. One of these is outliers or influential values. Especially in a small sample, these values can have a major influence on the modeling process. In the beta regression framework, this issue has been addressed mainly in two ways: the assumption of a different response model and the application of a minimum density power divergence estimation (MDPDE) procedure. In this paper, however, we propose a simple hierarchical Bayesian methodology in the context of a varying dispersion beta response model that is robust to outliers, as shown through an extensive simulation study and analysis of two real data sets. To robustify Bayesian modeling, a heavy-tailed Student's t prior with uniform degrees of freedom is adopted for the regression coefficients. This proposal results in a wieldy implementation procedure which avails practical use of the approach.
SiM Causal Inference Missing Data & Measurement Error Nonparametric & Semiparametric · 发表于 2018-11-22
Jing Qin, Tao Yu, Pengfei Li, Hao Liu, Baojiang Chen
摘要
The augmented inverse weighting method is one of the most popular methods for estimating the mean of the response in causal inference and missing data problems. An important component of this method is the propensity score. Popular parametric models for the propensity score include the logistic, probit, and complementary log‐log models. A common feature of these models is that the propensity score is a monotonic function of a linear combination of the explanatory variables. To avoid the need to choose a model, we model the propensity score via a semiparametric single‐index model, in which the score is an unknown monotonic nondecreasing function of the given single index. Under this new model, the augmented inverse weighting estimator (AIWE) of the mean of the response is asymptotically linear, semiparametrically efficient, and more robust than existing estimators. Moreover, we have made a surprising observation. The inverse probability weighting and AIWEs based on a correctly specified parametric model may have worse performance than their counterparts based on a nonparametric model. A heuristic explanation of this phenomenon is provided. A real‐data example is used to illustrate the proposed methods.
SMMR Diagnostics & Prediction Longitudinal Data · 发表于 2018-11-22
Loïc Ferrer, Hein Putter, Cécile Proust-Lima
摘要
After the diagnosis of a disease, one major objective is to predict cumulative probabilities of events such as clinical relapse or death from the individual information collected up to a prediction time, usually including biomarker repeated measurements. Several competing estimators have been proposed, mainly from two approaches: joint modelling and landmarking. These approaches differ by the information used, the model assumptions and the complexity of the computational procedures. This paper aims to review the two approaches, precisely define the derived estimators of dynamic predictions and compare their performances notably in case of misspecification. The ultimate goal is to provide key elements for the use of individual dynamic predictions in clinical practice. Prediction of two competing causes of prostate cancer progression from the history of prostate-specific antigen is used as a motivated example. We formally define the quantity to estimate and its estimators, propose techniques to assess the uncertainty around predictions and validate them. We then conduct an in-depth simulation study compare the estimators in terms of prediction error, discriminatory power, efficiency and robustness to model assumptions. We show that prediction tools should be handled with care, in particular by properly specifying models and estimators.
SiM Diagnostics & Prediction · 发表于 2018-11-22
Hui Xu, Jing Qian, Nina P. Paynter, Xuehong Zhang, Brian W. Whitcomb, Shelley S. Tworoger, Kathryn M. Rexrode, Susan E. Hankinson, Raji Balasubramanian
摘要
The matched case‐control design is frequently used in the study of complex disorders and can result in significant gains in efficiency, especially in the context of measuring biomarkers; however, risk prediction in this setting is not straightforward. We propose an inverse‐probability weighting approach to estimate the predictive ability associated with a set of covariates. In particular, we propose an algorithm for estimating the summary index, area under the curve corresponding to the Receiver Operating Characteristic curve associated with a set of pre‐defined covariates for predicting a binary outcome. By combining data from the parent cohort with that generated in a matched case control study, we describe methods for estimation of the population parameters of interest and the corresponding area under the curve. We evaluate the bias associated with the proposed methods in simulations by considering a range of parameter settings. We illustrate the methods in two data applications: (1) a prospective cohort study of cardiovascular disease in women, the Women's Health Study, and (2) a matched case‐control study nested within the Nurses' Health Study aimed at risk prediction of invasive breast cancer.
SiM Missing Data & Measurement Error Bayesian Methods Econometrics & Economics · 发表于 2018-11-22
Andrea Gabrio, Alexina J. Mason, Gianluca Baio
摘要
Economic evaluations from individual‐level data are an important component of the process of technology appraisal, with a view to informing resource allocation decisions. A critical problem in these analyses is that both effectiveness and cost data typically present some complexity (eg, nonnormality, spikes, and missingness) that should be addressed using appropriate methods. However, in routine analyses, standardised approaches are typically used, possibly leading to biassed inferences. We present a general Bayesian framework that can handle the complexity. We show the benefits of using our approach with a motivating example, the MenSS trial, for which there are spikes at one in the effectiveness and missingness in both outcomes. We contrast a set of increasingly complex models and perform sensitivity analysis to assess the robustness of the conclusions to a range of plausible missingness assumptions. We demonstrate the flexibility of our approach with a second example, the PBS trial, and extend the framework to accommodate the characteristics of the data in this study. This paper highlights the importance of adopting a comprehensive modelling approach to economic evaluations and the strategic advantages of building these complex models within a Bayesian framework.
SMMR Diagnostics & Prediction · 发表于 2018-11-20
Lei Huang, Jialiang Li
摘要
It is often necessary to differentiate subjects from multiple categories using medical tests. We may then adopt statistical measures to characterize the performance of these tests. The three-way ROC analysis has been proposed to evaluate the diagnostic accuracy of medical tests with three categories, reflecting the correct classification probabilities across all possible decision thresholds. The geometry of the ROC surface is carefully studied, leading to numerical summary measures such as the volume under the surface. This paper generalizes the global volume under the surface of three-way ROC analysis to the weighted volume under the surface (WVUS) by introducing a weight function emphasizing particular regions of correct classification probabilities. This generalization practically allows researchers to calculate the diagnostic accuracy for a medical or clinical biomarker while satisfactorily high probabilities of correct classification for one or two classes are conditionally ensured. We provide the asymptotic properties of the proposed nonparametric and parametric estimators of WVUS, which could easily lend support to statistical inferences. Some simulations have been conducted to assess the proposed estimators and also to demonstrate the necessity of WVUS. A real data analysis about liver cancer illustrates our methodology.
SiM Other · 发表于 2018-11-20
Sean McGrath, XiaoFei Zhao, Zhi Zhen Qin, Russell Steele, Andrea Benedetti
摘要
An aggregate data meta‐analysis is a statistical method that pools the summary statistics of several selected studies to estimate the outcome of interest. When considering a continuous outcome, typically each study must report the same measure of the outcome variable and its spread (eg, the sample mean and its standard error). However, some studies may instead report the median along with various measures of spread. Recently, the task of incorporating medians in meta‐analysis has been achieved by estimating the sample mean and its standard error from each study that reports a median in order to meta‐analyze the means. In this paper, we propose two alternative approaches to meta‐analyze data that instead rely on medians. We systematically compare these approaches via simulation study to each other and to methods that transform the study‐specific medians and spread into sample means and their standard errors. We demonstrate that the proposed median‐based approaches perform better than the transformation‐based approaches, especially when applied to skewed data and data with high inter‐study variance. Finally, we illustrate these approaches in a meta‐analysis of patient delay in tuberculosis diagnosis.
Biometrics Survival Analysis Nonparametric & Semiparametric · 发表于 2018-11-20
Meng Xia, Susan Murray, Nabihah Tayob
摘要
This article takes a fresh look at group sequential methods applied to two-sample tests of censored survival data and proposes an alternative method of defining and evaluating treatment benefit. Our method re-purposes traditional censored event time data into a sequence of short-term outcomes taken from (potentially overlapping) follow-up windows. A new two-sample restricted means test based on this restructured follow-up data is proposed along with group sequential methods for its use in the clinical trial setting. This method compares favorably with existing methods for group sequential monitoring of time-to-event outcomes, including methods for monitoring the restricted means test and the logrank test. Our method performs particularly well in cases where there is a delayed treatment effect and/or a subset of cured patients. As part of developing group sequential methods for these analyses, we consider asymmetric error spending approaches that differentially limit the chances of stopping incorrectly for perceived efficacy versus perceived harm attributed to the investigational arm of the trial. Recommendations for how to choose proper group sequential stopping boundaries are given, with supporting simulations and an example from the AIDS Clinical Trial Group.
SiM Other · 发表于 2018-11-20
A. James O'Malley, Peter James, Todd A. MacKenzie, Jinyoung Byun, S. V. Subramanian, Jason P. Block
摘要
Hierarchical modeling is the preferred approach of modeling neighborhood effects. When both residential and workplace neighborhoods are known, a bivariate (residential‐workplace) neighborhood random effect that quantifies the extent that a neighborhood's residential and workplace effects are correlated may be modeled. However, standard statistical software for hierarchical models does not easily allow correlations between the random effects of distinct clustering variables to be incorporated. To overcome this challenge, we develop a Bayesian model and an accompanying estimation procedure that allows for correlated bivariate neighborhood effects and allows individuals to reside or work in multiple neighborhoods, cross‐sectional and longitudinal heterogeneity between individuals, and serial correlation between repeated observations over time. Simulation studies that vary key model parameters evaluate how well each aspect of the model is identified by the data. We apply the model to the motivating Framingham Heart Study linked food establishment data to examine whether (i) proximity to fast‐food establishments is associated with body mass index, (ii) workplace neighborhood exposure associations are larger than those for residential neighborhood exposure, and (iii) residential neighborhood exposure associations correlate with workplace neighborhood exposure. Comparisons of the full model to models with restricted versions of the covariance structure illustrate the impact of including each feature of the covariance structure.
Biometrics High-Dimensional · 发表于 2018-11-20
Gen Li, Xiaokang Liu, Kun Chen
摘要
Multi-view data have been routinely collected in various fields of science and engineering. A general problem is to study the predictive association between multivariate responses and multi-view predictor sets, all of which can be of high dimensionality. It is likely that only a few views are relevant to prediction, and the predictors within each relevant view contribute to the prediction collectively rather than sparsely. We cast this new problem under the familiar multivariate regression framework and propose an integrative reduced-rank regression (iRRR), where each view has its own low-rank coefficient matrix. As such, latent features are extracted from each view in a supervised fashion. For model estimation, we develop a convex composite nuclear norm penalization approach, which admits an efficient algorithm via alternating direction method of multipliers. Extensions to non-Gaussian and incomplete data are discussed. Theoretically, we derive non-asymptotic oracle bounds of iRRR under a restricted eigenvalue condition. Our results recover oracle bounds of several special cases of iRRR including Lasso, group Lasso, and nuclear norm penalized regression. Therefore, iRRR seamlessly bridges group-sparse and low-rank methods and can achieve substantially faster convergence rate under realistic settings of multi-view learning. Simulation studies and an application in the Longitudinal Studies of Aging further showcase the efficacy of the proposed methods.
SiM Genetics & Omics · 发表于 2018-11-20
Tianzhong Yang, Han Chen, Hongwei Tang, Donghui Li, Peng Wei
摘要
As whole‐exome/genome sequencing data become increasingly available in genetic epidemiology research consortia, there is emerging interest in testing the interactions between rare genetic variants and environmental exposures that modify the risk of complex diseases. However, testing rare‐variant–based gene‐by‐environment interactions (GxE) is more challenging than testing the genetic main effects due to the difficulty in correctly estimating the latter under the null hypothesis of no GxE effects and the presence of neutral variants. In response, we have developed a family of powerful and data‐adaptive GxE tests, called “aGE” tests, in the framework of the adaptive powered score test, originally proposed for testing the genetic main effects. Using extensive simulations, we show that aGE tests can control the type I error rate in the presence of a large number of neutral variants or a nonlinear environmental main effect, and the power is more resilient to the inclusion of neutral variants than that of existing methods. We demonstrate the performance of the proposed aGE tests using Pancreatic Cancer Case‐Control Consortium Exome Chip data. An R package “aGE” is available at http://github.com/ytzhong/projects/.
SiM Diagnostics & Prediction Nonparametric & Semiparametric · 发表于 2018-11-20
Ting Lyu, Zhiliang Ying, Hong Zhang
摘要
When multiple biomarkers are available for disease diagnosis, it is desirable to efficiently combine them to form a single index. Making use of the Neyman‐Pearson paradigm, we propose a new combination/transformation approach to disease diagnosis that efficiently combines multiple biomarkers. The proposed method does not require that the biomarkers be jointly normally distributed or the covariance matrices for the diseased and the nondiseased are nondifferential. An R package is developed to implement the proposed method. Simulations and two real data examples demonstrate advantages of the new method over existing ones.
JASA Genetics & Omics High-Dimensional · 发表于 2018-11-19
Zijian Guo, Wanjie Wang, T. Tony Cai, Hongzhe Li
摘要
Estimating the genetic relatedness between two traits based on the genome-wide association data is an important problem in genetics research. In the framework of high-dimensional linear models, we introduce two measures of genetic relatedness and develop optimal estimators for them. One is genetic covariance, which is defined to be the inner product of the two regression vectors, and another is genetic correlation, which is a normalized inner product by their lengths. We propose functional de-biased estimators (FDEs), which consist of an initial estimation step with the plug-in scaled Lasso estimator, and a further bias correction step. We also develop estimators of the quadratic functionals of the regression vectors, which can be used to estimate the heritability of each trait. The estimators are shown to be minimax rate-optimal and can be efficiently implemented. Simulation results show that FDEs provide better estimates of the genetic relatedness than simple plug-in estimates. FDE is also applied to an analysis of a yeast segregant data set with multiple traits to estimate the genetic relatedness among these traits. *(abstract via PubMed)*