2021 年的期刊文章 · 第 12 页

共 1542 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2021-08-12 到 2021-08-01。

按期刊、按主题看 →

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 页

JRSSB Longitudinal Data Spatial & Time Series · 发表于 2021-08-12

Modelling the COVID-19 Infection Trajectory: A Piecewise Linear Quantile Trend Model

Feiyu Jiang, Zifeng Zhao, Xiaofeng Shao

摘要

We propose a piecewise linear quantile trend model to analyse the trajectory of the COVID-19 daily new cases (i.e. the infection curve) simultaneously across multiple quantiles. The model is intuitive, interpretable and naturally captures the phase transitions of the epidemic growth rate via change-points. Unlike the mean trend model and least squares estimation, our quantile-based approach is robust to outliers, captures heteroscedasticity (commonly exhibited by COVID-19 infection curves) and automatically delivers both point and interval forecasts with minimal assumptions. Building on a self-normalized (SN) test statistic, this paper proposes a novel segmentation algorithm for multiple change-point estimation. Theoretical guarantees such as segmentation consistency are established under mild and verifiable assumptions. Using the proposed method, we analyse the COVID-19 infection curves in 35 major countries and discover patterns with potentially relevant implications for effectiveness of the pandemic responses by different countries. A simple change-adaptive two-stage forecasting scheme is further designed to generate short-term prediction of COVID-19 cumulative new cases and is shown to deliver accurate forecast valuable to public health decision-making.

SiM Spatial & Time Series Latent Variables & Mixtures · 发表于 2021-08-11

Composite likelihood inference for ordinal periodontal data with replicated spatial patterns

Pingping Wang, Ting Fung Ma, Dipankar Bandyopadhyay, Yincai Tang, Jun Zhu

摘要

Spatial ordinal data observed separately for multiple subjects are common in biomedical research, yet statistical methodology for such ordinal data analysis is limited. The existing methodology often assumes a single realization of spatial ordinal data without replications, a commonplace in disease mapping studies, and thus are not directly applicable. Motivated by a dataset evaluating periodontal disease (PD) status, we propose a multisubject spatial ordinal model that assumes a geostatistical spatial structure within a regression framework through an elegant latent variable representation. For achieving computational scalability within a classical inferential framework, we develop a maximum composite likelihood method for parameter estimation, and establish the asymptotic properties of the parameter estimates. Another major contribution is the development of model diagnostic measures for our dependent data scenario using generalized surrogate residuals. A simulation study suggests sound finite sample properties of the proposed methods. We also illustrate our proposed methodology via application to the motivating PD dataset. A companion R package clordr is available for easy implementation.

SiM Causal Inference Missing Data & Measurement Error · 发表于 2021-08-10

Estimating adjusted risk differences by multiply‐imputing missing control binary potential outcomes following propensity score‐matching

Peter C. Austin, Donald B. Rubin, Neal Thomas

摘要

We describe a new method to combine propensity‐score matching with regression adjustment in treatment‐control studies when outcomes are binary by multiply imputing potential outcomes under control for the matched treated subjects. This enables the estimation of clinically meaningful measures of effect such as the risk difference. We used Monte Carlo simulation to explore the effect of the number of imputed potential outcomes under control for the matched treated subjects on inferences about the risk difference. We found that imputing potential outcomes under control (either single imputation or multiple imputation) resulted in a substantial reduction in bias compared with what was achieved using conventional nearest neighbor matching alone. Increasing the number of imputed potential outcomes under control resulted in more efficient estimation, with more efficient estimation of the estimated risk difference when increasing the number of the imputed potential outcomes. The greatest relative increase in efficiency was achieved by imputing five potential outcomes; once 20 outcomes under control were imputed for each matched treated subject, further improvements in efficiency were negligible. We also examined the effect of the number of these imputed potential outcomes on: (i) estimated standard errors; (ii) mean squared error; (iii) coverage of estimated confidence intervals. We illustrate the application of the method by estimating the effect on the risk of death within 1 year of prescribing beta‐blockers to patients discharged from hospital with a diagnosis of heart failure.

SiM Bayesian Methods · 发表于 2021-08-10

A multivariate statistical approach to predict COVID‐19 count data with epidemiological interpretation and uncertainty quantification

Francesco Bartolucci, Fulvia Pennoni, Antonietta Mira

摘要

For the analysis of COVID‐19 pandemic data, we propose Bayesian multinomial and Dirichlet‐multinomial autoregressive models for time‐series of counts of patients in mutually exclusive and exhaustive observational categories, defined according to the severity of the patient status and the required treatment. Categories include hospitalized in regular wards (H) and in intensive care units (ICU), together with deceased (D) and recovered (R). These models explicitly formulate assumptions on the transition probabilities between these categories across time, thanks to a flexible formulation based on parameters that a priori follow normal distributions, possibly truncated to incorporate specific hypotheses having an epidemiological interpretation. The posterior distribution of model parameters and the transition matrices are estimated by a Markov chain Monte Carlo algorithm that also provides predictions and allows us to compute the reproduction number . All estimates and predictions are endowed with an accuracy measure obtained thanks to the Bayesian approach. We present results concerning data collected during the first wave of the pandemic in Italy and Lombardy and study the effect of nonpharmaceutical interventions. Suitable discrepancy measures defined to check and compare models show that the Dirichlet‐multinomial model has an adequate fit and provides good predictive performance in particular for H and ICU patients.

SMMR Causal Inference · 发表于 2021-08-09

The change in estimate method for selecting confounders: A simulation study

Denis Talbot, Awa Diop, Mathilde Lavigne-Robichaud, Chantal Brisson

摘要

Background: The change in estimate is a popular approach for selecting confounders in epidemiology. It is recommended in epidemiologic textbooks and articles over significance test of coefficients, but concerns have been raised concerning its validity. Few simulation studies have been conducted to investigate its performance. Methods: An extensive simulation study was realized to compare different implementations of the change in estimate method. The implementations were also compared when estimating the association of body mass index with diastolic blood pressure in the PROspective Québec Study on Work and Health. Results: All methods were susceptible to introduce important bias and to produce confidence intervals that included the true effect much less often than expected in at least some scenarios. Overall mixed results were obtained regarding the accuracy of estimators, as measured by the mean squared error. No implementation adequately differentiated confounders from non-confounders. In the real data analysis, none of the implementation decreased the estimated standard error. Conclusion: Based on these results, it is questionable whether change in estimate methods are beneficial in general, considering their low ability to improve the precision of estimates without introducing bias and inability to yield valid confidence intervals or to identify true confounders.

JASA Spatial & Time Series High-Dimensional · 发表于 2021-08-09

Sparse Identification and Estimation of Large-Scale Vector AutoRegressive Moving Averages

Ines Wilms, Sumanta Basu, Jacob Bien, David S. Matteson

摘要

The Vector AutoRegressive Moving Average (VARMA) model is fundamental to the theory of multivariate time series; however, identifiability issues have led practitioners to abandon it in favor of the simpler but more restrictive Vector AutoRegressive (VAR) model. We narrow this gap with a new optimization-based approach to VARMA identification built upon the principle of parsimony. Among all equivalent data-generating models, we use convex optimization to seek the parameterization that is simplest in a certain sense. A user-specified strongly convex penalty is used to measure model simplicity, and that same penalty is then used to define an estimator that can be efficiently computed. We establish consistency of our estimators in a double-asymptotic regime. Our non-asymptotic error bound analysis accommodates both model specification and parameter estimation steps, a feature that is crucial for studying large-scale VARMA algorithms. Our analysis also provides new results on penalized estimation of infinite-order VAR, and elastic net regression under a singular covariance structure of regressors, which may be of independent interest. We illustrate the advantage of our method over VAR alternatives on three real data examples. *(abstract via PubMed)*

JASA Longitudinal Data Survey & Sampling · 发表于 2021-08-09

Simultaneous Inference for Empirical Best Predictors With a Poverty Study in Small Areas

Katarzyna Reluga, María-José Lombardía, Stefan Sperlich

摘要

Today, generalized linear mixed models (GLMM) are broadly used in many fields. However, the development of tools for performing simultaneous inference has been largely neglected in this domain. A framework for joint inference is indispensable to carry out statistically valid multiple comparisons of parameters of interest between all or several clusters. We therefore develop simultaneous confidence intervals and multiple testing procedures for empirical best predictors under GLMM. In addition, we implement our methodology to study widely employed examples of mixed models, that is, the unit-level binomial, the area-level Poisson-gamma and the area-level Poisson-lognormal mixed models. The asymptotic results are accompanied by extensive simulations. A case study on predicting poverty rates illustrates applicability and advantages of our simultaneous inference tools. *(abstract via OpenAlex)*

JASA Causal Inference Nonparametric & Semiparametric Missing Data & Measurement Error · 发表于 2021-08-09

Nonparametric Bounds for Causal Effects in Imperfect Randomized Experiments

Erin E. Gabriel, Arvid Sjölander, Michael C. Sachs

摘要

Nonignorable missingness and noncompliance can occur even in well-designed randomized experiments, making the intervention effect that the experiment was designed to estimate nonidentifiable. Nonparametric causal bounds provide a way to narrow the range of possible values for a nonidentifiable causal effect with minimal assumptions. We derive novel bounds for the causal risk difference for a binary outcome and intervention in randomized experiments with nonignorable missingness that is caused by a variety of mechanisms, with both perfect and imperfect compliance. We show that the so-called worst-case imputation, whereby all missing subjects on the intervention arm are assumed to have events and all missing subjects on the control or placebo arm are assumed to be event-free, can be too pessimistic in blinded studies with perfect compliance, and is not bounding the correct estimand with imperfect compliance. We illustrate the use of the proposed bounds in our motivating data example of peanut consumption on the development of peanut allergies in infants. We find that, even accounting for potentially nonignorable missingness and noncompliance, our derived bounds confirm that regular exposure to peanuts reduces the risk of development of peanut allergies, making the results of this study much more compelling. *(abstract via OpenAlex)*

JASA Other · 发表于 2021-08-09

Individualized Group Learning

Chencheng Cai, Rong Chen, Min-ge Xie

摘要

Many massive data sets are assembled through collections of information of a large number of individuals in a population. The analysis of such data, especially in the aspect of individualized inferences and solutions, has the potential to create significant value for practical applications. Traditionally, inference for an individual in the dataset is either solely relying on the information of the individual or from summarizing the information about the whole population. However, with the availability of big data, we have the opportunity, as well as a unique challenge, to make a more effective individualized inference that takes into consideration of both the population information and the individual discrepancy. To deal with the possible heterogeneity within the population while providing effective and credible inferences for individuals in a dataset, this article develops a new approach called the individualized group learning (iGroup). The iGroup approach uses local nonparametric techniques to generate an individualized group by pooling other entities in the population which share similar characteristics with the target individual, even when individual estimates are biased due to limited number of observations. Three general cases of iGroup are discussed, and their asymptotic performances are investigated. Both theoretical results and empirical simulations reveal that, by applying iGroup, the performance of statistical inference on the individual level are ensured and can be substantially improved from inference based on either solely individual information or entire population information. The method has a broad range of applications. An example in financial statistics is presented. *(abstract via OpenAlex)*

JASA Nonparametric & Semiparametric · 发表于 2021-08-09

Estimation of Knots in Linear Spline Models

Guangyu Yang, Baqun Zhang, Min Zhang

摘要

The linear spline model is able to accommodate nonlinear effects while allowing for an easy interpretation. It has significant applications in studying threshold effects and change-points. However, its application in practice has been limited by the lack of both rigorously studied and computationally convenient method for estimating knots. A key difficulty in estimating knots lies in the nondifferentiability. In this article, we study influence functions of regular and asymptotically linear estimators for linear spline models using the semiparametric theory. Based on the theoretical development, we propose a simple semismooth estimating equation approach to circumvent the nondifferentiability issue using modified derivatives, in contrast to the previous smoothing-based methods. Without relying on any smoothing parameters, the proposed method is computationally convenient. To further improve numerical stability, a two-step algorithm taking advantage of the analytic solution available when knots are known is developed to solve the proposed estimating equation. Consistency and asymptotic normality are rigorously derived using the empirical process theory. Simulation studies have shown that the two-step algorithm performs well in terms of both statistical and computational properties and improves over existing methods. Supplementary materials for this article are available online. *(abstract via OpenAlex)*

Biometrics Causal Inference · 发表于 2021-08-07

Weak‐instrument robust tests in two‐sample summary‐data Mendelian randomization

Sheng Wang, Hyunseung Kang

摘要

Mendelian randomization (MR) has been a popular method in genetic epidemiology to estimate the effect of an exposure on an outcome using genetic variants as instrumental variables (IV), with two‐sample summary‐data MR being the most popular. Unfortunately, instruments in MR studies are often weakly associated with the exposure, which can bias effect estimates and inflate Type I errors. In this work, we propose test statistics that are robust under weak‐instrument asymptotics by extending the Anderson–Rubin, Kleibergen, and the conditional likelihood ratio test in econometrics to two‐sample summary‐data MR. We also use the proposed Anderson–Rubin test to develop a point estimator and to detect invalid instruments. We conclude with a simulation and an empirical study and show that the proposed tests control size and have better power than existing methods with weak instruments.

Biometrics Survival Analysis Causal Inference Nonparametric & Semiparametric · 发表于 2021-08-07

On estimating optimal regime for treatment initiation time based on restricted mean residual lifetime

Xin Chen, Rui Song, Jiajia Zhang, Swann Arp Adams, Liuquan Sun, Wenbin Lu

摘要

When to initiate treatment on patients is an important problem in many medical studies such as AIDS and cancer. In this article, we formulate the treatment initiation time problem for time‐to‐event data and propose an optimal individualized regime that determines the best treatment initiation time for individual patients based on their characteristics. Different from existing optimal treatment regimes where treatments are undertaken at a pre‐specified time, here new challenges arise from the complicated missing mechanisms in treatment initiation time data and the continuous treatment rule in terms of initiation time. To tackle these challenges, we propose to use restricted mean residual lifetime as a value function to evaluate the performance of different treatment initiation regimes, and develop a nonparametric estimator for the value function, which is consistent even when treatment initiation times are not completely observable and their distribution is unknown. We also establish the asymptotic properties of the resulting estimator in the decision rule and its associated value function estimator. In particular, the asymptotic distribution of the estimated value function is nonstandard, which follows a weighted chi‐squared distribution. The finite‐sample performance of the proposed method is evaluated by simulation studies and is further illustrated with an application to a breast cancer data.

Biometrics Other · 发表于 2021-08-07

Analysis of local sensitivity to nonignorability with missing outcomes and predictors

Heng Chen, Daniel F. Heitjan

摘要

The ISNI (index of sensitivity to local nonignorability) method quantifies local sensitivity of parametric inferences to nonignorable missingness in an outcome variable. Here we extend ISNI to the situations where both outcomes and predictors can be missing and where the missingness mechanism can be either parametric or semi‐parametric. We define the quantity MinNI (minimum nonignorability) to be an approximation to the norm of the smallest value of the transformed nonignorability that gives a nonnegligible displacement of the estimate of the parameter of interest. We illustrate our method in a complete data set from which we synthetically delete observations according to various patterns. We then apply the method to real‐data examples involving the normal linear model and conditional logistic regression.

JRSSB Causal Inference Testing & Uncertainty · 发表于 2021-08-04

The Confidence Interval Method for Selecting Valid Instrumental Variables

Frank Windmeijer, Xiaoran Liang, Fernando P. Hartwig, Jack Bowden

摘要

We propose a new method, the confidence interval (CI) method, to select valid instruments from a larger set of potential instruments for instrumental variable (IV) estimation of the causal effect of an exposure on an outcome. Invalid instruments are such that they fail the exclusion conditions and enter the model as explanatory variables. The CI method is based on the CIs of the per instrument causal effects estimates and selects the largest group with all CIs overlapping with each other as the set of valid instruments. Under a plurality rule, we show that the resulting standard IV, or two-stage least squares (2SLS) estimator has oracle properties. This result is the same as for the hard thresholding with voting (HT) method of Guo et al. (Journal of the Royal Statistical Society : Series B, 2018, 80, 793–815). Unlike the HT method, the number of instruments selected as valid by the CI method is guaranteed to be monotonically decreasing for decreasing values of the tuning parameter. For the CI method, we can therefore use a downward testing procedure based on the Sargan (Econometrica, 1958, 26, 393–415) test for overidentifying restrictions and a main advantage of the CI downward testing method is that it selects the model with the largest number of instruments selected as valid that passes the Sargan test.

JRSSB Testing & Uncertainty 讨论 / 评论 · 发表于 2021-08-04

Priyantha Wijayatunga’s Contribution to the Discussion of ‘Gaussian Differential Privacy’ by Dong et al

Priyantha Wijayatunga

摘要

In privacy preserving, for example using a differential privacy framework, when a statistical operation is done on a database, then it is done so that its results are not overly dependent on any data record in the database. In this sense, one may think that this is all about avoiding, for example outliers, when the desired statistical results are generated. But on the other hand, utility of such a result is questionable. When an average of a certain measure is requested, giving the median or an average calculated from implementation of a suitable resampling procedure may preserve the privacy of individual data. Such methods may be simple to implement and may have an appreciable utility. Recently in Santos-Lozada et al. (2020) using US census data, the authors show that implementation of the differential privacy will produce dramatic changes in population counts for racial/ethnic minorities in small areas and less urban settings, significantly altering knowledge about health disparities in mortality. It is also important to note that, according to the Fundamental Law of Information Recovery; ‘overly accurate answers to too many questions (on statistics) will destroy privacy (of individual data) in a spectacular way’. The paper tries to use frequentist statistical hypothesis testing framework for defining their differential privacy framework. However, as many of us are aware the hypothesis testing is undergoing immense criticism, especially within the applied statistical community, for example so-called p-value problems. It may be that such problems and oppositions may appear in any privacy framework that is based on the frequentist statistical hypothesis testing methodology. Therefore, ideally the authors should touch upon such problems, especially to attract applied researchers (in the sense of the discipline of statistics) such as computer scientists, social scientists, etc., to their approach. In fact, the authors emphasize the importance of the use of the Neyman–Pearson hypothesis testing framework for interpreting differential privacy over other methods. According to original thesis of R. A. Fisher (1890–1962), the meaning of, for example, p-value < 0.05 is that the respective experiment should be repeated a few times. Such a practice should be handled by authors' composition results. Apart from above issues, it seems that some mathematical expressions should accompany with some verbal expressions too, for example the inequality in the Definition 1 is valid for a given (fixed) (S, S') pair. Therefore, it is helpful to write ℙ{M(S) ∈ E | S} rather than writing ℙ{M(S) ∈ E} and indicate how strong the conditional (in)dependence of M(S) on S' given S. Any other confusions should be eliminated. The author replied later, in writing, as follows. *(abstract via OpenAlex)*

JASA Bayesian Methods Survey & Sampling · 发表于 2021-08-04

Modeling the Marked Presence-Only Data: A Case Study of Estimating the Female Sex Worker Size in Malawi

Ian Laga, Xiaoyue Niu, Le Bao

摘要

Certain subpopulations like female sex workers (FSW), men who have sex with men (MSM), and people who inject drugs (PWID) often have higher prevalence of HIV/AIDS and are difficult to map directly due to stigma, discrimination, and criminalization. Fine-scale mapping of those populations contributes to the progress towards reducing the inequalities and ending the AIDS epidemic. In 2016 and 2017, the PLACE surveys were conducted at 3,290 venues in 20 out of the total 28 districts in Malawi to estimate the FSW sizes. These venues represent a presence-only data set where, instead of knowing both where people live and do not live (presence-absence data), only information about visited locations is available. In this study, we develop a Bayesian model for presence-only data and utilize the PLACE data to estimate the FSW size and uncertainty interval at a 1.5 × 1.5-km resolution for all of Malawi. The estimates can also be aggregated to any desirable level (city/district/region) for implementing targeted HIV prevention and treatment programs in FSW communities, which have been successful in lowering the incidence of HIV and other sexually transmitted infections. *(abstract via PubMed)*

JRSSB Testing & Uncertainty · 发表于 2021-08-04

Leveraging the Fisher Randomization Test using Confidence Distributions: Inference, Combination and Fusion Learning

Xiaokang Luo, Tirthankar Dasgupta, Minge Xie, Regina Y. Liu

摘要

The flexibility and wide applicability of the Fisher randomization test (FRT) make it an attractive tool for assessment of causal effects of interventions from modern-day randomized experiments that are increasing in size and complexity. This paper provides a theoretical inferential framework for FRT by establishing its connection with confidence distributions. Such a connection leads to development’s of (i) an unambiguous procedure for inversion of FRTs to generate confidence intervals with guaranteed coverage, (ii) new insights on the effect of size of the Monte Carlo sample on the estimation of a p-value curve and (iii) generic and specific methods to combine FRTs from multiple independent experiments with theoretical guarantees. Our developments pertain to finite sample settings but have direct extensions to large samples. Simulations and a case example demonstrate the benefit of these new developments.

JRSSB Longitudinal Data 讨论 / 评论 · 发表于 2021-08-04

Jorge Mateu’s Contribution to the Discussion of ‘Gaussian Differential Privacy’ by Dong et al

Jorge Mateu

摘要

The authors are to be congratulated on a valuable and thought-provoking contribution motivating this new framework for private data analysis, the f-differential privacy. A key aspect is the use of trade-off functions of hypothesis testing as a measure of indistinguishability of two or group neighbouring data sets. I would like to frame this contribution within the context of big data together with data mining and data science. As the authors point out, an increasing and unprecedented wealth of methods concern individual data recorded from personal devices or even private or public resources. In this large-scale and big data context, privacy in terms of anonymous personal information is key for any legal and serious data analysis. Modern data collection techniques allow tracking objects (persons) continuously. This means that we do not only know the current location of a moving object, but we also track the objects over time. A set of some tracks from different moving objects may be considered a trajectory pattern. Indeed, studying the behaviour of moving objects over time and their interaction, either between objects or with environment, plays a crucial role in understanding how they use space and more importantly how they interact with each other. In this context, a snapshot of a trajectory pattern might be seen as a spatial point pattern. I pose the following two cases. One is a data set where the individual events are themselves trajectories (functions) moving within a city. We need algorithms for privacy guarantee and to get groups of (trajectory) data anonymised. In a related context, assume the events are exact space–time coordinates of infected people from an infectious disease. So we have a spatiotemporal point pattern and we need to test if the ratio of the first-order intensity of the infected group against that of the control group behaves in a particular way. In other words, we need to test if groups of space–time events are distinguishable from other events. These two problems deal with data in space–time, and pose problems on privacy over space–time locations. I wonder if GDP applies over this spatial context. Also, a natural question is how f-DP or GDP can be used in contexts where type I and II errors are only approximated by simulations -because the probability distribution under the null or alternative hypothesis is usually unknown, as often happens in the spatiotemporal context. *(abstract via OpenAlex)*

JASA Networks & Graphs · 发表于 2021-08-04

Efficient Estimation for Random Dot Product Graphs via a One-Step Procedure

Fangzheng Xie, Yanxun Xu

摘要

We propose a one-step procedure to estimate the latent positions in random dot product graphs efficiently. Unlike the classical spectral-based methods, the proposed one-step procedure takes advantage of both the low-rank structure of the expected adjacency matrix and the Bernoulli likelihood information of the sampling model simultaneously. We show that for each vertex, the corresponding row of the one-step estimator (OSE) converges to a multivariate normal distribution after proper scaling and centering up to an orthogonal transformation, with an efficient covariance matrix. The initial estimator for the one-step procedure needs to satisfy the so-called approximate linearization property. The OSE improves the commonly adopted spectral embedding methods in the following sense: Globally for all vertices, it yields an asymptotic sum of squares error no greater than those of the spectral methods, and locally for each vertex, the asymptotic covariance matrix of the corresponding row of the OSE dominates those of the spectral embeddings in spectra. The usefulness of the proposed one-step procedure is demonstrated via numerical examples and the analysis of a real-world Wikipedia graph dataset. *(abstract via OpenAlex)*

JRSSB Testing & Uncertainty · 发表于 2021-08-04

Covariate Powered Cross-Weighted Multiple Testing

Nikolaos Ignatiadis, Wolfgang Huber

摘要

A fundamental task in the analysis of data sets with many variables is screening for associations. This can be cast as a multiple testing task, where the objective is achieving high detection power while controlling type I error. We consider m hypothesis tests represented by pairs ((Pi,Xi))1≤i≤m of p-values Pi and covariates Xi, such that Pi⊥Xi if Hi is null. Here, we show how to use information potentially available in the covariates about heterogeneities among hypotheses to increase power compared to conventional procedures that only use the Pi. To this end, we upgrade existing weighted multiple testing procedures through the independent hypothesis weighting (IHW) framework to use data-driven weights that are calculated as a function of the covariates. Finite sample guarantees, for example false discovery rate control, are derived from cross-weighting, a data-splitting approach that enables learning the weight-covariate function without overfitting as long as the hypotheses can be partitioned into independent folds, with arbitrary within-fold dependence. IHW has increased power compared to methods that do not use covariate information. A key implication of IHW is that hypothesis rejection in common multiple testing setups should not proceed according to the ranking of the p-values, but by an alternative ranking implied by the covariate-weighted p-values.

JRSSB Clinical Trials 讨论 / 评论 · 发表于 2021-08-03

Peter Krusche and Frank Bretz's Contribution to the Discussion of ‘Gaussian Differential Privacy’ by Dong et al

Peter Krusche, Frank Bretz

摘要

We congratulate Drs. Dong, Roth and Su for advancing our understanding of differential privacy from a hypothesis-testing perspective. As more personal data are collected for research, we need to mathematically understand when an adversary may be tempted to identify individuals and obtain sensitive information about them. This is especially true when working with patient data in healthcare, as respecting patient consent and privacy is imperative and subject to strong legal and regulatory constraints. In addition, not all data modalities collected in clinical trials are uniformly suitable for anonymization or de-identification. For example, only a few genetic markers or peripheral information present in medical images may be sufficient to uniquely re-identify individuals. As a result, patient privacy requirements can severely limit our ability to link data across clinical studies and build complex data sets to improve our understanding of disease and response to treatments. Despite the progress made so far, there is limited work on privacy-preserving techniques with applications to clinical research. When computing differentially private data summaries or model parameters on clinical data sets, we trade data utility off for privacy guarantees through randomized algorithms that are parameterized by a ‘privacy budget’. Data from clinical trials present both unique challenges and opportunities in this context: for one, knowledge of randomization by study design may allow us to suppress data with little impact on the summary being computed. For another, clinical trial data sets with only hundreds or thousands of participants are much smaller than other types of data typically considered in the context of differential privacy (insurance data, electronic health records, etc.), making it challenging to retain data utility as well as privacy. These specific constraints raise interesting questions as to how the proposed methods can be applied to clinical research, for example around the management of the differential privacy budget. As such, we believe it would be highly useful to study the practical implications of µ/f-DP and ε-DP for small clinical data sets. For example, could we address the trade-offs between data utility and preserving privacy in the framework of f-DP while acknowledging the customary hypothesis-testing approaches in clinical research where power is maximized while controlling the Type I error rate at a specified threshold. Moreover, we believe it would be interesting to study if there are non-trivial ways to utilize the rules of composition in f-DP for optimizing complex sets of private data summaries, or the training process of private generative models. Received in writing. *(abstract via OpenAlex)*

SiM Causal Inference Genetics & Omics · 发表于 2021-08-02

Testing and correcting for weak and pleiotropic instruments in two‐sample multivariable Mendelian randomization

Eleanor Sanderson, Wes Spiller, Jack Bowden

摘要

Multivariable Mendelian randomization (MVMR) is a form of instrumental variable analysis which estimates the direct effect of multiple exposures on an outcome using genetic variants as instruments. Mendelian randomization and MVMR are frequently conducted using two‐sample summary data where the association of the genetic variants with the exposures and outcome are obtained from separate samples. If the genetic variants are only weakly associated with the exposures either individually or conditionally, given the other exposures in the model, then standard inverse variance weighting will yield biased estimates for the effect of each exposure. Here, we develop a two‐sample conditional F ‐statistic to test whether the genetic variants strongly predict each exposure conditional on the other exposures included in a MVMR model. We show formally that this test is equivalent to the individual level data conditional F ‐statistic, indicating that conventional rule‐of‐thumb critical values of 10, can be used to test for weak instruments. We then demonstrate how reliable estimates of the causal effect of each exposure on the outcome can be obtained in the presence of weak instruments and pleiotropy, by repurposing a commonly used heterogeneity Q ‐statistic as an estimating equation. Furthermore, the minimized value of this Q ‐statistic yields an exact test for heterogeneity due to pleiotropy. We illustrate our methods with an application to estimate the causal effect of blood lipid fractions on age‐related macular degeneration.

SiM Causal Inference · 发表于 2021-08-02

Restricted sub‐tree learning to estimate an optimal dynamic treatment regime using observational data

Kelly Speth, Lu Wang

摘要

Dynamic treatment regimes (DTRs), consisting of a sequence of tailored treatment decision rules that span multiple stages of care, present a unique opportunity in our drive toward personalized medicine. Given that estimation of optimal DTRs is often exploratory and communication with clinicians is vital, robust and flexible methods that yield interpretable results are needed. Tree‐based methods utilizing a purity measure defined on the full set of covariates have enjoyed much success in meeting this goal. Often, however, it is necessary for clinical, practical, or ethical reasons to restrict certain covariates that should be used when making treatment decisions. Herein we present restricted sub‐tree learning (ReST‐L), a flexible and robust, sub‐tree‐based method to estimate an optimal multi‐stage multi‐treatment DTR that enables restrictions to the set of prespecified candidate tailoring variables. ReST‐L employs a purity measure derived from an augmented inverse probability weighted estimator for the counterfactual mean outcome, using observational data to build multi‐stage decision trees that are restricted in sub‐tree spaces defined by the corresponding prescriptive covariates. We show that ReST‐L is able to correctly estimate the optimal DTR searching over a large number of variables with relatively small sample sizes and improves upon competing estimation methods. We demonstrate the utility of ReST‐L to estimate a two‐stage fluid resuscitation strategy for patients admitted to an intensive care unit with acute emergent sepsis.

SiM Causal Inference · 发表于 2021-08-02

Pleiotropy robust methods for multivariable Mendelian randomization

Andrew J. Grant, Stephen Burgess

摘要

Mendelian randomization is a powerful tool for inferring the presence, or otherwise, of causal effects from observational data. However, the nature of genetic variants is such that pleiotropy remains a barrier to valid causal effect estimation. There are many options in the literature for pleiotropy robust methods when studying the effects of a single risk factor on an outcome. However, there are few pleiotropy robust methods in the multivariable setting, that is, when there are multiple risk factors of interest. In this article we introduce three methods which build on common approaches in the univariable setting: MVMR‐Robust; MVMR‐Median; and MVMR‐Lasso. We discuss the properties of each of these methods and examine their performance in comparison to existing approaches in a simulation study. MVMR‐Robust is shown to outperform existing outlier robust approaches when there are low levels of pleiotropy. MVMR‐Lasso provides the best estimation in terms of mean squared error for moderate to high levels of pleiotropy, and can provide valid inference in a three sample setting. MVMR‐Median performs well in terms of estimation across all scenarios considered, and provides valid inference up to a moderate level of pleiotropy. We demonstrate the methods in an applied example looking at the effects of intelligence, education and household income on the risk of Alzheimer's disease.

JRSSB Causal Inference High-Dimensional Testing & Uncertainty · 发表于 2021-08-02

Optimal Statistical Inference for Individualized Treatment Effects in High-Dimensional Models

Tianxi Cai, T. Tony Cai, Zijian Guo

摘要

The ability to predict individualized treatment effects (ITEs) based on a given patient's profile is essential for personalized medicine. We propose a hypothesis testing approach to choosing between two potential treatments for a given individual in the framework of high-dimensional linear models. The methodological novelty lies in the construction of a debiased estimator of the ITE and establishment of its asymptotic normality uniformly for an arbitrary future high-dimensional observation, while the existing methods can only handle certain specific forms of observations. We introduce a testing procedure with the type I error controlled and establish its asymptotic power. The proposed method can be extended to making inference for general linear contrasts, including both the average treatment effect and outcome prediction. We introduce the optimality framework for hypothesis testing from both the minimaxity and adaptivity perspectives and establish the optimality of the proposed procedure. An extension to high-dimensional approximate linear models is also considered. The finite sample performance of the procedure is demonstrated in simulation studies and further illustrated through an analysis of electronic health records data from patients with rheumatoid arthritis.

Biometrika Bayesian Methods · 发表于 2021-08-02

Estimation under matrix quadratic loss and matrix superharmonicity

T Matsuda, W E Strawderman

摘要

Summary: We investigate estimation of a normal mean matrix under the matrix quadratic loss. Improved estimation under the matrix quadratic loss implies improved estimation of any linear combination of the columns under the quadratic loss. First, an unbiased estimate of risk is derived and the Efron–Morris estimator is shown to be minimax. Next, a notion of matrix superharmonicity for matrix-variate functions is introduced and shown to have properties analogous to those of the usual superharmonic functions, which may be of independent interest. Then, it is shown that the generalized Bayes estimator with respect to a matrix superharmonic prior is minimax. We also provide a class of matrix superharmonic priors that includes the previously proposed generalization of Stein’s prior. Numerical results demonstrate that matrix superharmonic priors work well for low-rank matrices.

SiM Causal Inference · 发表于 2021-08-02

Estimating the marginal effect of a continuous exposure on an ordinal outcome using data subject to covariate‐driven treatment and visit processes

Janie Coulombe, Erica E. M. Moodie, Robert W. Platt

摘要

In the statistical literature, a number of methods have been proposed to ensure valid inference about marginal effects of variables on a longitudinal outcome in settings with irregular monitoring times. However, the potential biases due to covariate‐driven monitoring times and confounding have rarely been considered simultaneously, and never in a setting with an ordinal outcome and a continuous exposure. In this work, we propose and demonstrate a methodology for causal inference in such a setting, relying on a proportional odds model to study the effect of the exposure on the outcome. Irregular observation times are considered via a proportional rate model, and a generalization of inverse probability of treatment weights is used to account for the continuous exposure. We motivate our methodology by the estimation of the marginal (causal) effect of the time spent on video or computer games on suicide attempts in theAdd Healthstudy, a longitudinal study in the United States. Although in theAdd Healthdata, observation times are prespecified, our proposed approach is applicable even in more general settings such as when analyzing data from electronic health records where observations are highly irregular. In simulation studies, we let observation times vary across individuals and demonstrate that not accounting for biasing imbalances due to the monitoring and the exposure schemes can bias the estimate for the marginal odds ratio of exposure.

AoS Other · 发表于 2021-08-01

What is resolution? A statistical minimax testing perspective on superresolution microscopy

Gytis Kulaitis, Axel Munk, Frank Werner

摘要

As a general rule of thumb the resolution of a light microscope (i.e., the ability to discern objects) is predominantly described by the full width at half maximum (FWHM) of its point spread function (psf)—the diameter of the blurring density at half of its maximum. Classical wave optics suggests a linear relationship between FWHM and resolution also manifested in the well-known Abbe and Rayleigh criteria, dating back to the end of the 19th century. However, during the last two decades conventional light microscopy has undergone a shift from microscopic scales to nanoscales. This increase in resolution comes with the need to incorporate the random nature of observations (light photons) and challenges the classical view of discernability, as we argue in this paper. Instead, we suggest a statistical description of resolution obtained from such random data. Our notion of discernability is based on statistical testing whether one or two objects with the same total intensity are present. For Poisson measurements, we get linear dependence of the (minimax) detection boundary on the FWHM, whereas for a homogeneous Gaussian model the dependence of resolution is nonlinear. Hence, at small physical scales modeling by homogeneous gaussians is inadequate, although often implicitly assumed in many reconstruction algorithms. In contrast, the Poisson model and its variance stabilized Gaussian approximation seem to provide a statistically sound description of resolution at the nanoscale. Our theory is also applicable to other imaging setups, such as telescopes.

AoS Spatial & Time Series Nonparametric & Semiparametric · 发表于 2021-08-01

Volatility coupling

Jean Jacod, Jia Li, Zhipeng Liao

摘要

This paper provides a strong approximation, or coupling, theory for spot volatility estimators formed using high-frequency data. We show that the t-statistic process associated with the nonparametric spot volatility estimator can be strongly approximated by a growing-dimensional vector of independent variables defined as functions of Brownian increments. We use this coupling theory to study the uniform inference for the volatility process in an infill asymptotic setting. Specifically, we propose uniform confidence bands for spot volatility, beta, idiosyncratic variance processes, and their nonlinear transforms. The theory is also applied to address an open question concerning the inference of monotone nonsmooth integrated volatility functionals such as the occupation time and its quantiles.

AoS Bayesian Methods · 发表于 2021-08-01

Universal Bayes consistency in metric spaces

Steve Hanneke, Aryeh Kontorovich, Sivan Sabato, Roi Weiss

摘要

We extend a recently proposed 1-nearest-neighbor based multiclass learning algorithm and prove that our modification is universally strongly Bayes consistent in all metric spaces admitting any such learner, making it an “optimistically universal” Bayes-consistent learner. This is the first learning algorithm known to enjoy this property; by comparison, the k-NN classifier and its variants are not generally universally Bayes consistent, except under additional structural assumptions, such as an inner product, a norm, finite dimension or a Besicovitch-type property. The metric spaces in which universal Bayes consistency is possible are the “essentially separable” ones—a notion that we define, which is more general than standard separability. The existence of metric spaces that are not essentially separable is widely believed to be independent of the ZFC axioms of set theory. We prove that essential separability exactly characterizes the existence of a universal Bayes-consistent learner for the given metric space. In particular, this yields the first impossibility result for universal Bayes consistency. Taken together, our results completely characterize strong and weak universal Bayes consistency in metric spaces.

AoS High-Dimensional · 发表于 2021-08-01

The distribution of the Lasso: Uniform control over sparse balls and adaptive parameter tuning

Léo Miolane, Andrea Montanari

摘要

The Lasso is a popular regression method for high-dimensional problems in which the number of parameters θ1,…,θN, is larger than the number n of samples: N>n. A useful heuristics relates the statistical properties of the Lasso estimator to that of a simple soft-thresholding denoiser, in a denoising problem in which the parameters (θi)i≤N are observed in Gaussian noise, with a carefully tuned variance. Earlier work confirmed this picture in the limit n,N→∞, pointwise in the parameters θ and in the value of the regularization parameter. Here, we consider a standard random design model and prove exponential concentration of its empirical distribution around the prediction provided by the Gaussian denoising model. Crucially, our results are uniform with respect to θ belonging to ℓq balls, q∈[0,1], and with respect to the regularization parameter. This allows us to derive sharp results for the performances of various data-driven procedures to tune the regularization. Our proofs make use of Gaussian comparison inequalities, and in particular of a version of Gordon’s minimax theorem developed by Thrampoulidis, Oymak and Hassibi, which controls the optimum value of the Lasso optimization problem. Crucially, we prove a stability property of the minimizer in Wasserstein distance that allows one to characterize properties of the minimizer itself.

AoS Other · 发表于 2021-08-01

Stein’s method of normal approximation: Some recollections and reflections

Louis H. Y. Chen

摘要

This paper is a short exposition of Stein’s method of normal approximation from my personal perspective. It focuses mainly on the characterization of the normal distribution and the construction of Stein identities. Through examples, it provides glimpses into the many approaches to constructing Stein identities and the diverse applications of Stein’s method to mathematical problems. It also includes anecdotes of historical interest, including how Stein discovered his method and how I found an unpublished proof of his of the Berry–Esseen theorem.

AoS Nonparametric & Semiparametric · 发表于 2021-08-01

Stein 1956: Efficient nonparametric testing and estimation

A. W. van der Vaart, J. A. Wellner

AoS High-Dimensional · 发表于 2021-08-01

Second-order Stein: SURE for SURE and other applications in high-dimensional inference

Pierre C. Bellec, Cun-Hui Zhang

摘要

Stein’s formula states that a random variable of the form z⊤f(z)−divf(z) is mean-zero for all functions f with integrable gradient. Here, divf is the divergence of the function f and z is a standard normal vector. This paper aims to propose a second-order Stein formula to characterize the variance of such random variables for all functions f(z) with square integrable gradient, and to demonstrate the usefulness of this second-order Stein formula in various applications. In the Gaussian sequence model, a remarkable consequence of Stein’s formula is Stein’s Unbiased Risk Estimate (SURE), an unbiased estimate of the mean squared risk for almost any given estimator μˆ of the unknown mean vector. A first application of the second-order Stein formula is an Unbiased Risk Estimate for SURE itself (SURE for SURE): an unbiased estimate providing information about the squared distance between SURE and the squared estimation error of μˆ. SURE for SURE has a simple form as a function of the data and is applicable to all μˆ with square integrable gradient, for example, the Lasso and the Elastic Net. In addition to SURE for SURE, the following statistical applications are developed: (1) upper bounds on the risk of SURE when the estimation target is the mean squared error; (2) confidence regions based on SURE and using the second-order Stein formula; (3) oracle inequalities satisfied by SURE-tuned estimates under a mild Lipschtiz assumption; (4) an upper bound on the variance of the size of the model selected by the Lasso, and more generally an upper bound on the variance of the empirical degrees-of-freedom of convex penalized estimators; (5) explicit expressions of SURE for SURE for the Lasso and the Elastic Net; (6) in the linear model, a general semiparametric scheme to de-bias a differentiable initial estimator for the statistical inference of a low-dimensional projection of the unknown regression coefficient vector, with a characterization of the variance after debiasing; and (7) an accuracy analysis of a Gaussian Monte Carlo scheme to approximate the divergence of functions f:Rn→Rn.

AoS Latent Variables & Mixtures · 发表于 2021-08-01

Robust k-means clustering for distributions with two moments

Yegor Klochkov, Alexey Kroshnin, Nikita Zhivotovskiy

摘要

We consider the robust algorithms for the k-means clustering problem where a quantizer is constructed based on N independent observations. Our main results are median of means based nonasymptotic excess distortion bounds that hold under the two bounded moments assumption in a general separable Hilbert space. In particular, our results extend the renowned asymptotic result of (Ann. Statist. 9 (1981) 135–140) who showed that the existence of two moments is sufficient for strong consistency of an empirically optimal quantizer in Rd. In a special case of clustering in Rd, under two bounded moments, we prove matching (up to constant factors) nonasymptotic upper and lower bounds on the excess distortion, which depend on the probability mass of the lightest cluster of an optimal quantizer. Our bounds have the sub-Gaussian form, and the proofs are based on the versions of uniform bounds for robust mean estimators.

AoS Bayesian Methods Nonparametric & Semiparametric · 发表于 2021-08-01

Propriety of the reference posterior distribution in Gaussian process modeling

Joseph Muré

摘要

In a seminal article, Berger, De Oliveira and Sansó [J. Amer. Statist. Assoc. 96 (2001) 1361–1374] compare several objective prior distributions for the parameters of Gaussian process models with isotropic correlation kernel. The reference prior distribution stands out among them insofar as it always leads to a proper posterior. They prove this result for rough correlation kernels: Spherical, Exponential with power ρ<2, Matérn with smoothness ν<1. This paper provides a proof for smooth correlation kernels: Exponential with power ρ=2, Matérn with smoothness ν≥1, Rational Quadratic, along with tail rates of the reference prior for these kernels.

AoS Other · 发表于 2021-08-01

Peskun–Tierney ordering for Markovian Monte Carlo: Beyond the reversible scenario

Christophe Andrieu, Samuel Livingstone

摘要

Historically time-reversibility of the transitions or processes underpinning Markov chain Monte Carlo methods (MCMC) has played a key role in their development, while the self-adjointness of associated operators together with the use of classical functional analysis techniques on Hilbert spaces have led to powerful and practically successful tools to characterise and compare their performance. Similar results for algorithms relying on nonreversible Markov processes are scarce. We show that for a type of nonreversible Monte Carlo Markov chains and processes, of current or renewed interest in the physics and statistical literatures, it is possible to develop comparison results which closely mirror those available in the reversible scenario. We show that these results shed light on earlier literature, proving some conjectures and strengthening some earlier results.

AoS Spatial & Time Series Testing & Uncertainty · 发表于 2021-08-01

Optimal change-point estimation in time series

Ngai Hang Chan, Wai Leong Ng, Chun Yip Yau, Haihan Yu

摘要

This paper establishes asymptotic theory for optimal estimation of change points in general time series models under α-mixing conditions. We show that the Bayes-type estimator is asymptotically minimax for change-point estimation under squared error loss. Two bootstrap procedures are developed to construct confidence intervals for the change points. An approximate limiting distribution of the change-point estimator under small change is also derived. Simulations and real data applications are presented to investigate the finite sample performance of the Bayes-type estimator and the bootstrap procedures.

AoS Machine Learning Networks & Graphs Clinical Trials · 发表于 2021-08-01

On the rate of convergence of fully connected deep neural network regression estimates

Michael Kohler, Sophie Langer

摘要

Recent results in nonparametric regression show that deep learning, that is, neural network estimates with many hidden layers, are able to circumvent the so-called curse of dimensionality in case that suitable restrictions on the structure of the regression function hold. One key feature of the neural networks used in these results is that their network architecture has a further constraint, namely the network sparsity. In this paper, we show that we can get similar results also for least squares estimates based on simple fully connected neural networks with ReLU activation functions. Here, either the number of neurons per hidden layer is fixed and the number of hidden layers tends to infinity suitably fast for sample size tending to infinity, or the number of hidden layers is bounded by some logarithmic factor in the sample size and the number of neurons per hidden layer tends to infinity suitably fast for sample size tending to infinity. The proof is based on new approximation results concerning deep neural networks.

AoS Bayesian Methods · 发表于 2021-08-01

On extended admissible procedures and their nonstandard Bayes risk

Haosui Duanmu, Daniel M. Roy

摘要

For finite parameter spaces, among decision procedures with finite risk functions, a decision procedure is extended admissible if and only if it is Bayes. Various relaxations of this classical equivalence have been established for infinite parameter spaces, but these extensions are each subject to technical conditions that limit their applicability, especially to modern (semi and nonparametric) statistical problems. Using results in mathematical logic and nonstandard analysis, we extend this equivalence to arbitrary statistical decision problems: informally, we show that, among decision procedures with finite risk functions, a decision procedure is extended admissible if and only if it has infinitesimal excess Bayes risk. In contrast to existing results, our equivalence holds in complete generality, that is, without regularity conditions or restrictions on the model or loss function. We also derive a nonstandard analogue of Blyth’s method that yields sufficient conditions for admissibility, and apply the nonstandard theory to derive a purely standard theorem: when risk functions are continuous on a compact Hausdorff parameter space, a procedure is extended admissible if and only if it is Bayes.

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 页