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 27 页
共 1294 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2018-01-22 到 2018。
按期刊、按主题看 →
Biometrics Spatial & Time Series · 发表于 2018-01-22
Stefano Castruccio, Hernando Ombao, Marc G. Genton
摘要
Summary: Functional Magnetic Resonance Imaging (fMRI) is a primary modality for studying brain activity. Modeling spatial dependence of imaging data at different spatial scales is one of the main challenges of contemporary neuroimaging, and it could allow for accurate testing for significance in neural activity. The high dimensionality of this type of data (on the order of hundreds of thousands of voxels) poses serious modeling challenges and considerable computational constraints. For the sake of feasibility, standard models typically reduce dimensionality by modeling covariance among regions of interest (ROIs)—coarser or larger spatial units—rather than among voxels. However, ignoring spatial dependence at different scales could drastically reduce our ability to detect activation patterns in the brain and hence produce misleading results. We introduce a multi-resolution spatio-temporal model and a computationally efficient methodology to estimate cognitive control related activation and whole-brain connectivity. The proposed model allows for testing voxel-specific activation while accounting for non-stationary local spatial dependence within anatomically defined ROIs, as well as regional dependence (between-ROIs). The model is used in a motor-task fMRI study to investigate brain activation and connectivity patterns aimed at identifying associations between these patterns and regaining motor functionality following a stroke.
Biometrics Meta-Analysis & Data Integration · 发表于 2018-01-22
Alexander M. Kaizer, Brian P. Hobbs, Joseph S. Koopmeiners
摘要
Summary: Traditional paradigms for clinical translation are challenged in settings where multiple contemporaneous therapeutic strategies have been identified as potentially beneficial. Platform trials have emerged as an approach for sequentially comparing multiple trials using a single protocol. The Ebola virus disease outbreak in West Africa represents one recent example which utilized a platform design. Specifically, the PREVAIL II master protocol sequentially tested new combinations of therapies against the concurrent, optimal standard of care (oSOC) strategy. Once a treatment demonstrated sufficient evidence of benefit, the treatment was added to the oSOC for all future comparisons (denoted as segments throughout the manuscript). In the interest of avoiding bias stemming from population drift, PREVAIL II considered only within-segment comparisons between the oSOC and novel treatments and failed to leverage data from oSOC patients in prior segments. This article describes adaptive design methodology aimed at boosting statistical power through Bayesian modeling and adaptive randomization. Specifically, the design uses multi-source exchangeability models to combine data from multiple segments and adaptive randomization to achieve information balance within a segment. When compared to the PREVAIL II design, we demonstrate that our proposed adaptive platform design improves power by as much as 51% with limited type-I error inflation. Further, the adaptive platform effectuates more balance with respect to the distribution of acquired information among study arms, with more patients randomized to experimental regimens.
Biometrika Spatial & Time Series · 发表于 2018-01-20
K Fokianos, M Pitsillou
摘要
SUMMARYWe introduce the matrix multivariate auto-distance covariance and correlation functions for time series, discuss their interpretation and develop consistent estimators for practical implementation. We also develop a test of the independent and identically distributed hypothesis for multivariate time series data and show that it performs better than the multivariate Ljung–Box test. We discuss computational aspects and present a data example to illustrate the method. *(abstract via OpenAlex)*
Biometrika Bayesian Methods · 发表于 2018-01-20
Wentao Li, Paul Fearnhead
摘要
Many statistical applications involve models for which it is difficult to evaluate the likelihood, but from which it is relatively easy to sample. Approximate Bayesian computation is a likelihood-free method for implementing Bayesian inference in such cases. We present results on the asymptotic variance of estimators obtained using approximate Bayesian computation in a large data limit. Our key assumption is that the data are summarized by a fixed-dimensional summary statistic that obeys a central limit theorem. We prove asymptotic normality of the mean of the approximate Bayesian computation posterior. This result also shows that, in terms of asymptotic variance, we should use a summary statistic that is of the same dimension as the parameter vector, |$p$|, and that any summary statistic of higher dimension can be reduced, through a linear transformation, to dimension |$p$| in a way that can only reduce the asymptotic variance of the posterior mean. We look at how the Monte Carlo error of an importance sampling algorithm that samples from the approximate Bayesian computation posterior affects the accuracy of estimators. We give conditions on the importance sampling proposal distribution such that the variance of the estimator will be of the same order as that of the maximum likelihood estimator based on the summary statistics used. This suggests an iterative importance sampling algorithm, which we evaluate empirically on a stochastic volatility model. *(abstract via OpenAlex)*
Biometrika Bayesian Methods High-Dimensional Spatial & Time Series · 发表于 2018-01-19
Jian Kang, Brian J Reich, Ana-Maria Staicu
摘要
This work concerns spatial variable selection for scalar-on-image regression. We propose a new class of Bayesian nonparametric models and develop an efficient posterior computational aigorithm. The proposed soft-thresholded Gaussian process provides large prior support over the class of piecewise-smooth, sparse, and continuous spatially-varying regression coefficient functions. In addition, under some mild regularity conditions the soft-thresholded Gaussian proess prior leads to the posterior consistency for parameter estimation and variable selection for scalar-on-image regression, even when the number of predictors is larger than the sample size. The proposed method is compared to alternatives via simulation and applied to an electroen-cephalography study of alcoholism. *(abstract via PubMed)*
Biometrika Nonparametric & Semiparametric · 发表于 2018-01-19
R H Spady, S Stouli
摘要
We propose dual regression as an alternative to quantile regression for the global estimation of conditional distribution functions. Dual regression provides the interpretational power of quantile regression while avoiding the need to repair intersecting conditional quantile surfaces. We introduce a mathematical programming characterization of conditional distribution functions which, in its simplest form, is the dual program of a simultaneous estimator for linear location-scale models, and use it to specify and estimate a flexible class of conditional distribution functions. We present asymptotic theory for the corresponding empirical dual regression process. *(abstract via OpenAlex)*
SiM Meta-Analysis & Data Integration · 发表于 2018-01-18
Tim P. Morris, David J. Fisher, Michael G. Kenward, James R. Carpenter
摘要
Quantitative evidence synthesis through meta‐analysis is central to evidence‐based medicine. For well‐documented reasons, the meta‐analysis of individual patient data is held in higher regard than aggregate data. With access to individual patient data, the analysis is not restricted to a “two‐stage” approach (combining estimates and standard errors) but can estimate parameters of interest by fitting a single model to all of the data, a so‐called “one‐stage” analysis. There has been debate about the merits of one‐ and two‐stage analysis. Arguments for one‐stage analysis have typically noted that a wider range of models can be fitted and overall estimates may be more precise. The two‐stage side has emphasised that the models that can be fitted in two stages are sufficient to answer the relevant questions, with less scope for mistakes because there are fewer modelling choices to be made in the two‐stage approach. For Gaussian data, we consider the statistical arguments for flexibility and precision in small‐sample settings. Regarding flexibility, several of the models that can be fitted only in one stage may not be of serious interest to most meta‐analysis practitioners. Regarding precision, we consider fixed‐ and random‐effects meta‐analysis and see that, for a model making certain assumptions, the number of stages used to fit this model is irrelevant; the precision will be approximately equal. Meta‐analysts should choose modelling assumptions carefully. Sometimes relevant models can only be fitted in one stage. Otherwise, meta‐analysts are free to use whichever procedure is most convenient to fit the identified model.
SiM Nonparametric & Semiparametric Survival Analysis · 发表于 2018-01-17
Tianlei Chen, Pang Du
摘要
Survival data with a cured portion are commonly seen in clinical trials. Motivated from a biological interpretation of cancer metastasis, promotion time cure model is a popular alternative to the mixture cure rate model for analyzing such data. The existing promotion cure models all assume a restrictive parametric form of covariate effects, which can be incorrectly specified especially at the exploratory stage. In this paper, we propose a nonparametric approach to modeling the covariate effects under the framework of promotion time cure model. The covariate effect function is estimated by smoothing splines via the optimization of a penalized profile likelihood. Point‐wise interval estimates are also derived from the Bayesian interpretation of the penalized profile likelihood. Asymptotic convergence rates are established for the proposed estimates. Simulations show excellent performance of the proposed nonparametric method, which is then applied to a melanoma study.
SMMR Missing Data & Measurement Error Causal Inference · 发表于 2018-01-17
Ying Zhang, Todd A Alonzo, for the Alzheimer’s Disease Neuroimaging Initiative
摘要
The receiver-operating characteristic surface is frequently used for presenting the accuracy of a diagnostic test for three-category classification problems. One common problem that can complicate the estimation of the volume under receiver-operating characteristic surface is that not all subjects receive the verification of the true disease status. Estimation based only on data from subjects with verified disease status may be biased, which is referred to as verification bias. In this article, we propose new verification bias correction methods to estimate the volume under receiver-operating characteristic surface for a continuous diagnostic test. We assume the verification process is missing not at random, which means the missingness might be related to unobserved clinical characteristics. Three classes of estimators are proposed, namely, inverse probability weighted, imputation-based, and doubly robust estimators. A jackknife estimator of variance is derived for all the proposed volume under receiver-operating characteristic surface estimators. The finite sample properties of the new estimators are examined via simulation studies. We illustrate our methods with data collected from Alzheimer’s disease research.
SMMR Genetics & Omics Survival Analysis · 发表于 2018-01-16
Wei Wei, Zequn Sun, Willian A da Silveira, Zhenning Yu, Andrew Lawson, Gary Hardiman, Linda E Kelemen, Dongjun Chung
摘要
Identification of cancer patient subgroups using high throughput genomic data is of critical importance to clinicians and scientists because it can offer opportunities for more personalized treatment and overlapping treatments of cancers. In spite of tremendous efforts, this problem still remains challenging because of low reproducibility and instability of identified cancer subgroups and molecular features. In order to address this challenge, we developed Integrative Genomics Robust iDentification of cancer subgroups (InGRiD), a statistical approach that integrates information from biological pathway databases with high-throughput genomic data to improve the robustness for identification and interpretation of molecularly-defined subgroups of cancer patients. We applied InGRiD to the gene expression data of high-grade serous ovarian cancer from The Cancer Genome Atlas and the Australian Ovarian Cancer Study. The results indicate clear benefits of the pathway-level approaches over the gene-level approaches. In addition, using the proposed InGRiD framework, we also investigate and address the issue of gene sharing among pathways, which often occurs in practice, to further facilitate biological interpretation of key molecular features associated with cancer progression. The R package “InGRiD” implementing the proposed approach is currently available in our research group GitHub webpage ( https://dongjunchung.github.io/INGRID/ ).
SMMR Other · 发表于 2018-01-16
Tsung-Shan Tsou
摘要
We construct a legitimate likelihood function for the agreement kappa coefficient for correlated data without specifically modelling all levels of correlation. This makes available the likelihood ratio test, the score test and other tools without the knowledge of the underlying distributions. This parametric robust likelihood approach applies to general clustered data scenarios. We provide simulations and real data analysis to demonstrate the advantage of the robust procedure.
SiM Missing Data & Measurement Error · 发表于 2018-01-15
Benedict H.W. Wong, Sarah B. Peskoe, Donna Spiegelman
摘要
The partial population attributable risk (pPAR) is used to quantify the population‐level impact of preventive interventions in a multifactorial disease setting. In this paper, we consider the effect of nondifferential risk factor misclassification on the direction and magnitude of bias of pPAR estimands and related quantities. We found that the bias in the uncorrected pPAR depends nonlinearly and nonmonotonically on the sensitivities, specificities, relative risks, and joint prevalence of the exposure of interest and background risk factors, as well as the associations between these factors. The bias in the uncorrected pPAR is most dependent on the sensitivity of the exposure. The magnitude of bias varies over a large range, and in a small region of the parameter space determining the pPAR, the direction of bias is away from the null. In contrast, the crude PAR can only be unbiased or biased towards the null by risk factor misclassification. The semiadjusted PAR is calculated using the formula for the crude PAR but plugs in the multivariate‐adjusted relative risk. Because the crude and semiadjusted PARs continue to be used in public health research, we also investigated the magnitude and direction of the bias that may arise when using these formulae instead of the pPAR. These PAR estimators and their uncorrected counterparts were calculated in a study of risk factors for colorectal cancer in the Health Professionals Follow‐up Study, where it was found that because of misclassification, the pPAR for low folate intake was overestimated with a relative bias of 48%, when red meat and alcohol intake were treated as misclassified risk factors that are not modified, and when red meat was treated as the modifiable risk factor, the estimated value of the pPAR went from 14% to 60%, further illustrating the extent to which misclassification can bias estimates of the pPAR.
SiM Missing Data & Measurement Error Latent Variables & Mixtures Longitudinal Data · 发表于 2018-01-15
Yongqiang Tang
摘要
The controlled imputation method refers to a class of pattern mixture models that have been commonly used as sensitivity analyses of longitudinal clinical trials with nonignorable dropout in recent years. These pattern mixture models assume that participants in the experimental arm after dropout have similar response profiles to the control participants or have worse outcomes than otherwise similar participants who remain on the experimental treatment. In spite of its popularity, the controlled imputation has not been formally developed for longitudinal binary and ordinal outcomes partially due to the lack of a natural multivariate distribution for such endpoints. In this paper, we propose 2 approaches for implementing the controlled imputation for binary and ordinal data based respectively on the sequential logistic regression and the multivariate probit model. Efficient Markov chain Monte Carlo algorithms are developed for missing data imputation by using the monotone data augmentation technique for the sequential logistic regression and a parameter‐expanded monotone data augmentation scheme for the multivariate probit model. We assess the performance of the proposed procedures by simulation and the analysis of a schizophrenia clinical trial and compare them with the fully conditional specification, last observation carried forward, and baseline observation carried forward imputation methods.
SMMR Latent Variables & Mixtures Survival Analysis Bayesian Methods · 发表于 2018-01-15
Jingheng Cai, Chenyi Liang
摘要
Respiratory cancer is one of the most commonly diagnosed cancers as well as the leading cause of cancer death. Numerous efforts have been devoted to reducing the death rate of respiratory cancer. In this article, we propose a semi-parametric Cox model with latent variables to assess the effects of observed and latent risk factors on survival time of respiratory cancer. The characteristics of latent risk factors are characterized via multiple observed indicators by a confirmatory factor analysis model. We develop a Bayesian estimation procedure to obtain the estimates of parameters. Simulation shows that the performance of the proposed methodology is satisfactory. The proposed method is applied to analyze the Surveillance, Epidemiology, and End Results Program data set.
SMMR Spatial & Time Series Longitudinal Data · 发表于 2018-01-15
Zhuqing Liu, Andreas J Bartsch, Veronica J Berrocal, Timothy D Johnson
摘要
Spatial resolution plays an important role in functional magnetic resonance imaging studies as the signal-to-noise ratio increases linearly with voxel volume. In scientific studies, where functional magnetic resonance imaging is widely used, the standard spatial resolution typically used is relatively low which ensures a relatively high signal-to-noise ratio. However, for pre-surgical functional magnetic resonance imaging analysis, where spatial accuracy is paramount, high-resolution functional magnetic resonance imaging may play an important role with its greater spatial resolution. High spatial resolution comes at the cost of a smaller signal-to-noise ratio. This begs the question as to whether we can leverage the higher signal-to-noise ratio of a standard functional magnetic resonance imaging study with the greater spatial accuracy of a high-resolution functional magnetic resonance imaging study in a pre-operative patient. To answer this question, we propose to regress the statistic image from a high resolution scan onto the statistic image obtained from a standard resolution scan using a mixed-effects model with spatially varying coefficients. We evaluate our model via simulation studies and we compare its performance with a recently proposed model that operates at a single spatial resolution. We apply and compare the two models on data from a patient awaiting tumor resection. Both simulation study results and the real data analysis demonstrate that our newly proposed model indeed leverages the larger signal-to-noise ratio of the standard spatial resolution scan while maintaining the advantages of the high spatial resolution scan.
SMMR Bayesian Methods · 发表于 2018-01-12
Camilla Bianchi, Ettore Lanzarone, Giustina Casagrande, Maria Laura Costantino
摘要
Hemodialysis is the most common therapy to treat renal insufficiency. However, notwithstanding the recent improvements, hemodialysis is still associated with a non-negligible rate of comorbidities, which could be reduced by customizing the treatment. Many differential compartment models have been developed to describe the mass balance of blood electrolytes and catabolites during hemodialysis, with the goal of improving and controlling hemodialysis sessions. However, these models often refer to an average uremic patient, while on the contrary the clinical need for customization requires patient-specific models. In this work, we assume that the customization can be obtained by means of patient-specific model parameters. We propose and validate a Bayesian approach to estimate the patient-specific parameters of a multi-compartment model, and to predict the single patient’s response to the treatment, in order to prevent intra-dialysis complications. The likelihood function is obtained by means of a discretized version of the multi-compartment model, where the discretization is in terms of a Runge–Kutta method to guarantee convergence, and the posterior densities of model parameters are obtained through Markov Chain Monte Carlo simulation. Results show fair estimations and the applicability in the clinical practice.
SiM Causal Inference Testing & Uncertainty · 发表于 2018-01-10
Edward Bein, Jonah Deutsch, Guanglei Hong, Kristin E. Porter, Xu Qin, Cheng Yang
摘要
This study investigates appropriate estimation of estimator variability in the context of causal mediation analysis that employs propensity score‐based weighting. Such an analysis decomposes the total effect of a treatment on the outcome into an indirect effect transmitted through a focal mediator and a direct effect bypassing the mediator. Ratio‐of‐mediator‐probability weighting estimates these causal effects by adjusting for the confounding impact of a large number of pretreatment covariates through propensity score‐based weighting. In step 1, a propensity score model is estimated. In step 2, the causal effects of interest are estimated using weights derived from the prior step's regression coefficient estimates. Statistical inferences obtained from this 2‐step estimation procedure are potentially problematic if the estimated standard errors of the causal effect estimates do not reflect the sampling uncertainty in the estimation of the weights. This study extends to ratio‐of‐mediator‐probability weighting analysis a solution to the 2‐step estimation problem by stacking the score functions from both steps. We derive the asymptotic variance‐covariance matrix for the indirect effect and direct effect 2‐step estimators, provide simulation results, and illustrate with an application study. Our simulation results indicate that the sampling uncertainty in the estimated weights should not be ignored. The standard error estimation using the stacking procedure offers a viable alternative to bootstrap standard error estimation. We discuss broad implications of this approach for causal analysis involving propensity score‐based weighting.
SiM Nonparametric & Semiparametric Testing & Uncertainty · 发表于 2018-01-10
Günter Heimann, Rossella Belleli, Jouni Kerman, Roland Fisch, Joseph Kahn, Sigrid Behr, Conny Berlin
摘要
Signal detection is routinely applied to spontaneous report safety databases in the pharmaceutical industry and by regulators. As an example, methods that search for increases in the frequencies of known adverse drug reactions for a given drug are routinely applied, and the results are reported to the health authorities on a regular basis. Such methods need to be sensitive to detect true signals even when some of the adverse drug reactions are rare. The methods need to be specific and account for multiplicity to avoid false positive signals when the list of known adverse drug reactions is long. To apply them as part of a routine process, the methods also have to cope with very diverse drugs (increasing or decreasing number of cases over time, seasonal patterns, very safe drugs versus drugs for life‐threatening diseases). In this paper, we develop new nonparametric signal detection methods, directed at detecting differences between a reporting and a reference period, or trends within a reporting period. These methods are based on bootstrap and permutation distributions, and they combine statistical significance with clinical relevance. We conducted a large simulation study to understand the operating characteristics of the methods. Our simulations show that the new methods have good power and control the family‐wise error rate at the specified level. Overall, in all scenarios that we explored, the method performs much better than our current standard in terms of power, and it generates considerably less false positive signals as compared to the current standard.
SiM Missing Data & Measurement Error · 发表于 2018-01-09
Paul W. Bernhardt
摘要
Missing covariate values are prevalent in regression applications. While an array of methods have been developed for estimating parameters in regression models with missing covariate data for a variety of response types, minimal focus has been given to validation of the response model and influence diagnostics. Previous research has mainly focused on estimating residuals for observations with missing covariates using expected values, after which specialized techniques are needed to conduct proper inference. We suggest a multiple imputation strategy that allows for the use of standard methods for residual analyses on the imputed data sets or a stacked data set. We demonstrate the suggested multiple imputation method by analyzing the Sleep in Mammals data in the context of a linear regression model and the New York Social Indicators Status data with a logistic regression model.
SMMR Survival Analysis Nonparametric & Semiparametric · 发表于 2018-01-09
MinJae Lee, Mohammad H Rahbar, Hooshang Talebi
摘要
We propose a nonparametric test for interactions when we are concerned with investigation of the simultaneous effects of two or more factors in a median regression model with right censored survival data. Our approach is developed to detect interaction in special situations, when the covariates have a finite number of levels with a limited number of observations in each level, and it allows varying levels of variance and censorship at different levels of the covariates. Through simulation studies, we compare the power of detecting an interaction between the study group variable and a covariate using our proposed procedure with that of the Cox Proportional Hazard (PH) model and censored quantile regression model. We also assess the impact of censoring rate and type on the standard error of the estimators of parameters. Finally, we illustrate application of our proposed method to real life data from Prospective Observational Multicenter Major Trauma Transfusion (PROMMTT) study to test an interaction effect between type of injury and study sites using median time for a trauma patient to receive three units of red blood cells. The results from simulation studies indicate that our procedure performs better than both Cox PH model and censored quantile regression model based on statistical power for detecting the interaction, especially when the number of observations is small. It is also relatively less sensitive to censoring rates or even the presence of conditionally independent censoring that is conditional on the levels of covariates.
SiM Clinical Trials Causal Inference Latent Variables & Mixtures · 发表于 2018-01-08
Karla Hemming, Monica Taljaard, Andrew Forbes
摘要
Cluster randomized trials are frequently used in health service evaluation. It is common practice to use an analysis model with a random effect to allow for clustering at the analysis stage. In designs where clusters are exposed to both control and treatment conditions, it may be of interest to examine treatment effect heterogeneity across clusters. In designs where clusters are not exposed to both control and treatment conditions, it can also be of interest to allow heterogeneity in the degree of clustering between arms. These two types of heterogeneity are related. It has been proposed in both parallel cluster trials, stepped‐wedge, and other cross‐over designs that this heterogeneity can be allowed for by incorporating additional random effect(s) into the model. Here, we show that the choice of model parameterization needs careful consideration as some parameterizations for additional heterogeneity induce unnecessary or implausible assumptions. We suggest more appropriate parameterizations, discuss their relative advantages, and demonstrate the implications of these model choices using a real example of a parallel cluster trial and a simulated stepped‐wedge trial.
SiM Other · 发表于 2018-01-08
Dan Jackson, Martin Law, Theo Stijnen, Wolfgang Viechtbauer, Ian R. White
摘要
Comparative trials that report binary outcome data are commonly pooled in systematic reviews and meta‐analyses. This type of data can be presented as a series of 2‐by‐2 tables. The pooled odds ratio is often presented as the outcome of primary interest in the resulting meta‐analysis. We examine the use of 7 models for random‐effects meta‐analyses that have been proposed for this purpose. The first of these models is the conventional one that uses normal within‐study approximations and a 2‐stage approach. The other models are generalised linear mixed models that perform the analysis in 1 stage and have the potential to provide more accurate inference. We explore the implications of using these 7 models in the context of a Cochrane Review, and we also perform a simulation study. We conclude that generalised linear mixed models can result in better statistical inference than the conventional 2‐stage approach but also that this type of model presents issues and difficulties. These challenges include more demanding numerical methods and determining the best way to model study specific baseline risks. One possible approach for analysts is to specify a primary model prior to performing the systematic review but also to present the results using other models in a sensitivity analysis. Only one of the models that we investigate is found to perform poorly so that any of the other models could be considered for either the primary or the sensitivity analysis.
Biometrika Optimization & Learning Theory · 发表于 2018-01-04
L Hong, T A Kuffner, R Martin
摘要
In a regression context, when the relevant subset of explanatory variables is uncertain, it is common to use a data-driven model selection procedure. Classical linear model theory, applied naively to the selected submodel, may not be valid because it ignores the selected submodel’s dependence on the data. We provide an explanation of this phenomenon, in terms of overfitting, for a class of model selection criteria. *(abstract via OpenAlex)*
Biometrika Other · 发表于 2018-01-04
N A Heard, P Rubin-Delanchy
摘要
Combining |$p$|-values from independent statistical tests is a popular approach to meta-analysis, particularly when the data underlying the tests are either no longer available or are difficult to combine. Numerous |$p$|-value combination methods appear in the literature, each with different statistical properties, yet often the final choice used in a meta-analysis can seem arbitrary, as if all effort has been expended in building the models that gave rise to the |$p$|-values. Birnbaum (1954) showed that any reasonable |$p$|-value combiner must be optimal against some alternative hypothesis. Starting from this perspective and recasting each method of combining |$p$|-values as a likelihood ratio test, we present theoretical results for some standard combiners that provide guidance on how a powerful combiner might be chosen in practice. *(abstract via OpenAlex)*
Biometrika Econometrics & Economics · 发表于 2018-01-03
N Lunardon
摘要
Firth (1993) introduced a method for reducing the bias of the maximum likelihood estimator. Here it is shown that the approach is also effective in reducing the sensitivity of inferential procedures to incidental parameters. *(abstract via OpenAlex)*
SiM Other · 发表于 2018-01-03
Kristoffer H. Hellton, Nils Lid Hjort
摘要
Statistical prediction methods typically require some form of fine‐tuning of tuning parameter(s), withK‐fold cross‐validation as the canonical procedure. For ridge regression, there exist numerous procedures, but common for all, including cross‐validation, is that one single parameter is chosen for all future predictions. We propose instead to calculate a unique tuning parameter for each individual for which we wish to predict an outcome. This generates an individualized prediction by focusing on the vector of covariates of a specific individual. The focused ridge—fridge—procedure is introduced with a 2‐part contribution: First we define an oracle tuning parameter minimizing the mean squared prediction error of a specific covariate vector, and then we propose to estimate this tuning parameter by using plug‐in estimates of the regression coefficients and error variance parameter. The procedure is extended to logistic ridge regression by using parametric bootstrap. For high‐dimensional data, we propose to use ridge regression with cross‐validation as the plug‐in estimate, and simulations show that fridge gives smaller average prediction error than ridge with cross‐validation for both simulated and real data. We illustrate the new concept for both linear and logistic regression models in 2 applications of personalized medicine: predicting individual risk and treatment response based on gene expression data. The method is implemented in the R packagefridge.
JMLR Spatial & Time Series · 发表于 2018
Emmanuel Bacry, Martin Bompaire, Philip Deegan, Stéphane Gaïffas, Søren V. Poulsen
摘要
This paper introduces tick, is a statistical learning library for Python 3, with a particular emphasis on time-dependent models, such as point processes, tools for generalized linear models and survival analysis. The core of the library provides model computational classes, solvers and proximal operators for regularization. It relies on a C++ implementation and state- of-the-art optimization algorithms to provide very fast computations in a single node multi-core setting. Source code and documentation can be downloaded from https://github.com/X-DataInitiative/tick.
JMLR Other · 发表于 2018
Jacob Schreiber
摘要
We present pomegranate, an open source machine learning package for probabilistic modeling in Python. Probabilistic modeling encompasses a wide range of methods that explicitly describe uncertainty using probability distributions. Three widely used probabilistic models implemented in pomegranate are general mixture models, hidden Markov models, and Bayesian networks. A primary focus of pomegranate is to abstract away the complexities of training models from their definition. This allows users to focus on specifying the correct model for their application instead of being limited by their understanding of the underlying algorithms. An aspect of this focus involves the collection of additive sufficient statistics from data sets as a strategy for training models. This approach trivially enables many useful learning strategies, such as out-of-core learning, minibatch learning, and semi-supervised learning, without requiring the user to consider how to partition data or modify the algorithms to handle these tasks themselves. pomegranate is written in Cython to speed up calculations and releases the global interpreter lock to allow for built-in multithreaded parallelism, making it competitive with---or outperform---other implementations of similar algorithms. This paper presents an overview of the design choices in pomegranate, and how they have enabled complex features to be supported by simple code. The code is available at \url{https://github.com/jmschrei/pomegranate}
JMLR Machine Learning Networks & Graphs · 发表于 2018
Michael Freitag, Shahin Amiriparian, Sergey Pugachevskiy, Nicholas Cummins, Björn Schuller
摘要
auDeep is a Python toolkit for deep unsupervised representation learning from acoustic data. It is based on a recurrent sequence to sequence autoencoder approach which can learn representations of time series data by taking into account their temporal dynamics. We provide an extensive command line interface in addition to a Python API for users and developers, both of which are comprehensively documented and publicly available at https://github.com/auDeep/auDeep. Experimental results indicate that auDeep features are competitive with state-of-the art audio classification.
Econometrica Causal Inference · 发表于 2018
Toru Kitagawa, Aleksey Tetenov
摘要
One of the main objectives of empirical analysis of experiments and quasi-experiments is to inform policy decisions that determine the allocation of treatments to individuals with different observable covariates. We study the properties and implementation of the Empirical Welfare Maximization (EWM) method, which estimates a treatment assignment policy by maximizing the sample analog of average social welfare over a class of candidate treatment policies. The EWM approach is attractive in terms of both statistical performance and practical implementation in realistic settings of policy design. Common features of these settings include: (i) feasible treatment assignment rules are constrained exogenously for ethical, legislative, or political reasons, (ii) a policy maker wants a simple treatment assignment rule based on one or more eligibility scores in order to reduce the dimensionality of individual observable characteristics, and/or (iii) the proportion of individuals who can receive the treatment is a priori limited due to a budget or a capacity constraint. We show that when the propensity score is known, the average social welfare attained by EWM rules converges at least at n1/2 rate to the maximum obtainable welfare uniformly over a minimally constrained class of data distributions, and this uniform convergence rate is minimax optimal. We examine how the uniform convergence rate depends on the richness of the class of candidate decision rules, the distribution of conditional treatment effects, and the lack of knowledge of the propensity score. We offer easily implementable algorithms for computing the EWM rule and an application using experimental data from the National JTPA Study. *(abstract via OpenAlex)*
JMLR Optimization & Learning Theory · 发表于 2018
Jiyan Yang, Yin-Lam Chow, Christopher Ré, Michael W. Mahoney
摘要
In recent years, stochastic gradient descent (SGD) methods and randomized linear algebra (RLA) algorithms have been applied to many large-scale problems in machine learning and data analysis. SGD methods are easy to implement and applicable to a wide range of convex optimization problems. In contrast, RLA algorithms provide much stronger performance guarantees but are applicable to a narrower class of problems. We aim to bridge the gap between these two methods in solving constrained overdetermined linear regression problems---e.g., $\ell_2$ and $\ell_1$ regression problems. We propose a hybrid algorithm named PWSGD that uses RLA techniques for preconditioning and constructing an importance sampling distribution, and then performs an SGD-like iterative process with weighted sampling on the preconditioned system. By rewriting a deterministic $\ell_p$ regression problem as a stochastic optimization problem, we connect PWSGD to several existing $\ell_p$ solvers including RLA methods with algorithmic leveraging (RLA for short). We prove that PWSGD inherits faster convergence rates that only depend on the lower dimension of the linear system, while maintaining low computation complexity. Such SGD convergence rates are superior to other related SGD algorithm such as the weighted randomized Kaczmarz algorithm. Particularly, when solving $\ell_1$ regression with size $n$ by $d$, PWSGD returns an approximate solution with $\epsilon$ relative error in the objective value in $\mathcal{O}(\log n \cdot \text{nnz}(A) + \text{poly}(d)/\epsilon^2)$ time. This complexity is uniformly better than that of RLA methods in terms of both $\epsilon$ and $d$ when the problem is unconstrained. In the presence of constraints, PWSGD only has to solve a sequence of much simpler and smaller optimization problem over the same constraints. In general this is more efficient than solving the constrained subproblem required in RLA. For $\ell_2$ regression, PWSGD returns an approximate solution with $\epsilon$ relative error in the objective value and the solution vector measured in prediction norm in $\mathcal{O}(\log n \cdot \text{nnz}(A) + \text{poly}(d) \log(1/\epsilon) /\epsilon)$ time. We show that for unconstrained $\ell_2$ regression, this complexity is comparable to that of RLA and is asymptotically better over several state-of-the-art solvers in the regime where the desired accuracy $\epsilon$, high dimension $n$ and low dimension $d$ satisfy $d\geq 1/\epsilon$ and $n \geq d^2/\epsilon$. We also provide lower bounds on the coreset complexity for more general regression problems, indicating that still new ideas will be needed to extend similar RLA preconditioning ideas to weighted SGD algorithms for more general regression problems. Finally, the effectiveness of such algorithms is illustrated numerically on both synthetic and real datasets, and the results are consistent with our theoretical findings and demonstrate that PWSGD converges to a medium- precision solution, e.g., $\epsilon=10^{-3}$, more quickly.
JMLR Bayesian Methods · 发表于 2018
James Hensman, Nicolas Durrande, Arno Solin
摘要
This work brings together two powerful concepts in Gaussian processes: the variational approach to sparse approximation and the spectral representation of Gaussian processes. This gives rise to an approximation that inherits the benefits of the variational approach but with the representational power and computational scalability of spectral representations. The work hinges on a key result that there exist spectral features related to a finite domain of the Gaussian process which exhibit almost-independent covariances. We derive these expressions for Matérn kernels in one dimension, and generalize to more dimensions using kernels with specific structures. Under the assumption of additive Gaussian noise, our method requires only a single pass through the data set, making for very fast and accurate computation. We fit a model to 4 million training points in just a few minutes on a standard laptop. With non- conjugate likelihoods, our MCMC scheme reduces the cost of computation from $\mathcal{O}(NM^2)$ (for a sparse Gaussian process) to $\mathcal{O}(NM)$ per iteration, where $N$ is the number of data and $M$ is the number of features.
JMLR Other · 发表于 2018
Kai-Yang Chiang, Inderjit S. Dhillon, Cho-Jui Hsieh
摘要
Learning a low-rank matrix from missing and corrupted observations is a fundamental problem in many machine learning applications. However, the role of side information in low-rank matrix learning has received little attention, and most current approaches are either ad-hoc or only applicable in certain restrictive cases. In this paper, we propose a general model that exploits side information to better learn low-rank matrices from missing and corrupted observations, and show that the proposed model can be further applied to several popular scenarios such as matrix completion and robust PCA. Furthermore, we study the effect of side information on sample complexity and show that by using our model, the efficiency for learning can be improved given sufficiently informative side information. This result thus provides theoretical insight into the usefulness of side information in our model. Finally, we conduct comprehensive experiments in three real-world applications---relationship prediction, semi-supervised clustering and noisy image classification, showing that our proposed model is able to properly exploit side information for more effective learning both in theory and practice.
Econometrica Causal Inference Clinical Trials · 发表于 2018
Magne Mogstad, Andres Santos, Alexander Torgovitsky
摘要
We propose a method for using instrumental variables (IV) to draw inference about causal effects for individuals other than those affected by the instrument at hand. Policy relevance and external validity turn on the ability to do this reliably. Our method exploits the insight that both the IV estimand and many treatment parameters can be expressed as weighted averages of the same underlying marginal treatment effects. Since the weights are identified, knowledge of the IV estimand generally places some restrictions on the unknown marginal treatment effects, and hence on the values of the treatment parameters of interest. We show how to extract information about the treatment parameter of interest from the IV estimand and, more generally, from a class of IV‐like estimands that includes the two stage least squares and ordinary least squares estimands, among others. Our method has several applications. First, it can be used to construct nonparametric bounds on the average causal effect of a hypothetical policy change. Second, our method allows the researcher to flexibly incorporate shape restrictions and parametric assumptions, thereby enabling extrapolation of the average effects for compliers to the average effects for different or larger populations. Third, our method can be used to test model specification and hypotheses about behavior, such as no selection bias and/or no selection on gain. *(abstract via OpenAlex)*
Econometrica Other · 发表于 2018
Paul Heidhues, Botond Kőszegi, Philipp Strack
摘要
We explore the learning process and behavior of an individual with unrealistically high expectations (“overconfidence”) when outcomes also depend on an external fundamental that affects the optimal action. Moving beyond existing results in the literature, we show that the agent’s beliefs regarding the fundamental converge under weak conditions. Furthermore, we identify a broad class of situations in which “learning” about the fundamental is self-defeating: it leads the individual systematically away from the correct belief and toward lower performance. Due to his overconfidence, the agent—even if initially correct—becomes too pessimistic about the fundamental. As he adjusts his behavior in response, he lowers outcomes and hence becomes even more pessimistic about the fundamental, perpetuating the misdirected learning. The greater is the loss from choosing a suboptimal action, the further the agent’s action ends up from optimal. We partially characterize environments in which self-defeating learning occurs, and show that the decisionmaker learns to take the optimal action if and only if a specific non-identifiability condition is satisfied. In contrast to an overconfident agent, an underconfident agent’s misdirected learning is self-limiting and therefore not very harmful. We argue that the decision situations in question are common in economic settings, including delegation, organizational, effort, and public-policy choices. *(abstract via OpenAlex)*
Econometrica Causal Inference · 发表于 2018
James J. Heckman, Rodrigo Pinto
摘要
This paper defines and analyzes a new monotonicity condition for the identification of counterfactuals and treatment effects in unordered discrete choice models with multiple treatments, heterogenous agents and discrete-valued instruments. Unordered monotonicity implies and is implied by additive separability of choice of treatment equations in terms of observed and unobserved variables. These results follow from properties of binary matrices developed in this paper. We investigate conditions under which unordered monotonicity arises as a consequence of choice behavior. We characterize IV estimators of counterfactuals as solutions to discrete mixture problems. *(abstract via PubMed)*
JMLR Other · 发表于 2018
Lyudmila Grigoryeva, Juan-Pablo Ortega
摘要
A new class of non-homogeneous state-affine systems is introduced for use in reservoir computing. Sufficient conditions are identified that guarantee first, that the associated reservoir computers with linear readouts are causal, time-invariant, and satisfy the fading memory property and second, that a subset of this class is universal in the category of fading memory filters with stochastic almost surely uniformly bounded inputs. This means that any discrete-time filter that satisfies the fading memory property with random inputs of that type can be uniformly approximated by elements in the non-homogeneous state-affine family.
JMLR Causal Inference Spatial & Time Series Nonparametric & Semiparametric · 发表于 2018
Massil Achab, Emmanuel Bacry, Stéphane Gaïffas, Iacopo Mastromatteo, Jean-François Muzy
摘要
We design a new nonparametric method that allows one to estimate the matrix of integrated kernels of a multivariate Hawkes process. This matrix not only encodes the mutual influences of each node of the process, but also disentangles the causality relationships between them. Our approach is the first that leads to an estimation of this matrix without any parametric modeling and estimation of the kernels themselves. As a consequence, it can give an estimation of causality relationships between nodes (or users), based on their activity timestamps (on a social network for instance), without knowing or estimating the shape of the activities lifetime. For that purpose, we introduce a moment matching method that fits the second-order and the third-order integrated cumulants of the process. A theoretical analysis allows us to prove that this new estimation technique is consistent. Moreover, we show, on numerical experiments, that our approach is indeed very robust with respect to the shape of the kernels and gives appealing results on the MemeTracker database and on financial order book data.