首页期刊导航|Journal of the royal statistical society, Series C. Applied statistics
期刊信息/Journal information
Journal of the royal statistical society, Series C. Applied statistics
Royal Statistical Society
Journal of the royal statistical society, Series C. Applied statistics

Royal Statistical Society

0035-9254

Journal of the royal statistical society, Series C. Applied statistics/Journal Journal of the royal statistical society, Series C. Applied statistics
正式出版
收录年代

    A Bayesian spatio-temporal Poisson auto-regressive model for the disease infection rate: application to COVID-19 cases in England

    Pierfrancesco Alaimo Di LoroDankmar BoehningSujit K. Sahu
    551-575页
    查看更多>>摘要：The COVID-19 pandemic provided new modelling challenges to investigate epidemic processes. This paper extends Poisson auto-regression to incorporate spatio-temporal dependence and characterize the local dynamics by borrowing information from adjacent areas. Adopted in a fully Bayesian framework and implemented through a novel sparse-matrix representation in Stan, the model has been validated through a simulation study. We use it to analyse the weekly COVID-19 cases in the English local authority districts and verify some of the epidemic-driving factors. The model detects substantial spatio-temporal heterogeneity and enables the formalization of novel model-based investigation methods for assessing additional aspects of disease epidemiology.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Exploring first and second-order spatio-temporal structures of lightning strike impacts in the French Alps using heavy subsampling

    Juliette BlanchetJean-Frangois CoeurjollyAlexis Pellerin
    576-597页
    查看更多>>摘要：This applied spatial statistics paper deals with a dataset of cloud-to-ground lightning strike impacts in the French Alps over the period 2011-2021 (approximately 1.4 million of events) modelled by a spatio-temporal point process. We explore first and higher-order structure for this point pattern and address the questions of homogeneity of the intensity function, first-order separability and dependence between events. Due to the nature of the dataset (very inhomogeneous in space and time) and the large amount of data, most of the nonparametric methods and statistical tests we consider lead to numerical problems or exceed clusters timeout. We suggest different subsampling strategies strongly reducing the number of events to overcome these difficulties and show how they can be used to draw conclusions on the initial point pattern.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Combining density forecast accuracy tests: an application to agricultural, energy, and metal commodities

    Bernardina AlgieriArturo LeccaditoDanilo SicoliDiana Tunaru...
    598-616页
    查看更多>>摘要：This study develops a new methodology for combining density forecast accuracy tests and assessing the relevance of psychological indicators in predicting commodity returns. Density forecasts provide a complete description of the uncertainty associated with a prediction and are highly requested by policymakers, central bankers, and financial operators to define policy actions, manage financial risks, and assess portfolio selection. The proposed methodology combines different tests and derives the p-value of the resulting test statistic by Monte Carlo simulations. To assess the power of the proposed methodology, we implement a set of experiments for several data-generating processes. Based on an empirical forecasting exercise applied to agricultural, energy, and metal commodities, we find that sentiment variables and psychological factors improve the density forecasts of commodity futures returns, especially for agricultural commodities. Additionally, combinations of sentiment variables are more powerful in predicting returns than considering them separately.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Personalized dynamic super learning: an application in predicting hemodiafiltration convection volumes

    Arthur ChattonMichele BallyRenee LevesqueIvana Malenica...
    617-637页
    查看更多>>摘要：Obtaining continuously updated predictions is a major challenge for personalized medicine. Leveraging combinations of parametric regressions and machine learning algorithms, the personalized online super learner (POSL) can achieve such dynamic and personalized predictions. We adapt POSL to predict a repeated continuous outcome dynamically and propose a new way to validate such personalized or dynamic prediction models. We illustrate its performance by predicting the convection volume of patients undergoing hemodiafiltration. POSL outperformed its candidate learners with respect to median absolute error, calibration-in-the-large, discrimination, and net benefit. We finally discuss the choices and challenges underlying the use of POSL.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Understanding power grid network vulnerability through the stochastic lens of network motif evolution

    Yuzhou ChenHon Keung Tony NgYulia R. GelH. Vincent Poor...
    638-658页
    查看更多>>摘要：Modern cyber-physical systems must exhibit high reliability since their failures can lead to catastrophic cascading events. Enhancing our understanding of the mechanisms behind the functionality of such networks is a key to ensuring the resilience of many critical infrastructures. In this paper, we develop a novel stochastic model, based on topological measures of complex networks, as a framework within which to examine such functionality. The key idea is to evaluate the dynamics of network motifs as descriptors of the underlying network topology and its response to adverse events. Our experiments on multiple power grid networks show that the proposed approach offers a new competitive pathway for resilience quantification of complex systems.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Evaluating treatment effect modifiers using data from randomized two-sequence, two-period crossover clinical trials: application to a diabetes study

    Ryo EmotoMasataka IgetaKota MatsuiKiyoaki Ishii...
    659-674页
    查看更多>>摘要：Type 2 diabetes is known as a heterogeneous disease with diverse pathophysiology. The identification of treatment effect modifiers for personalized medicine, however, has been difficult in basic research due to complex biological mechanisms related to various genetic and nongenetic factors, and in clinical research due to confounding bias and limited sample sizes. In this paper, we focus on a two-sequence, two-period crossover (CO) clinical trial of type 2 diabetes with baseline markers as a new strategy for analyzing treatment effect modification that allows confounding elimination and efficiency enhancement by within-patient treatment comparison. We provide a framework for statistical analysis of treatment effect modification and develop methods for testing for treatment effect modification and estimating individualized predictors of treatment differences in the CO trial with limited sample size. Numerical assessments showed that the efficiency of the proposed CO analysis was substantially higher than the standard analysis based on parallel-group comparison in these analyses. Application to the diabetes trial showed that markers based on gene expression in peripheral blood cells before pharmacological treatment significantly modified the effect of the treatment on the response and that marker-based predictors produced two subgroups of patients for whom different treatments should be recommended.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Wastewater surveillance using differentiate Gaussian processes

    Emily SomersetPatrick E. Brown
    675-690页
    查看更多>>摘要：Wastewater-based surveillance tracks disease spread within communities by analyzing biological markers in wastewater. A key component of effective wastewater-based surveillance is the reliable inference of underlying viral signals and their changes for accurate interpretation and dissemination. This paper proposes a Bayesian hierarchical modelling framework to jointly estimate wastewater viral signals and their derivatives, while accounting for common features and limitations of wastewater data. Our framework uses differentiable Gaussian processes to model both a common viral trend and deviations at individual stations. Specifically, the common trend is modelled as an Integrated Wiener Process and station-specific signals are smoothed assuming a Matern covariance function of order 1.5. We demonstrate the framework's utility by modelling SARS-CoV-2 concentrations across Canada and London, UK, as well as pepper mild mottle virus-normalized respiratory syncytial virus concentrations in Central California. Our results show that this framework reliably estimates both the signal and its derivative in retrospective and surveillance contexts, and show that inference of the signal's average rates of change is sensitive to the differentiability of the modelling process.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Fast spatial simulation of extreme high-resolution radar precipitation data using integrated nested Laplace approximations

    Silius M. VandeskogRaphaeel HuserOddbjorn BrulandSara Martino...
    691-716页
    查看更多>>摘要：Aiming to deliver improved precipitation simulations for hydrological impact assessment studies, we develop a methodology for modelling and simulating high-dimensional spatial precipitation extremes, focusing on both their marginal distributions and tail dependence structures. Tail dependence is crucial for assessing the consequences of extreme precipitation events, yet most stochastic weather generators do not attempt to capture this property. The spatial distribution of precipitation occurrences is modelled with four competing models, while the spatial distribution of nonzero extreme precipitation intensities are modelled with a latent Gaussian version of the spatial conditional extremes model. Nonzero precipitation marginal distributions are modelled using latent Gaussian models with gamma and generalized Pareto likelihoods. Fast inference is achieved using integrated nested Laplace approximations. We model and simulate spatial precipitation extremes in Central Norway, using 13 years of hourly radar data with a spatial resolution of 1 x 1 km2, over an area of size 6,461 km2, to describe the behaviour of extreme precipitation over a small drainage area. Inference on this high-dimensional data set is achieved within hours, and the simulations capture the main trends of the observed precipitation well.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Bayesian state-space models for the modelling and prediction of the results of English Premier League football

    P. Gareth RidallAndrew C. TitmanAnthony N. Pettitt
    717-742页
    查看更多>>摘要：The attraction of using state-space models (SSMs) is their ability to efficiently and dynamically predict in the presence of change. In this paper, we formulate a Bayesian SSM capable of predicting the outcomes of football matches and the associated states, which are the attacking and defensive strengths of each side and the common home goal advantage. Our filter achieves accuracy and efficiency by exploiting conjugacy in its update step and using exact expressions to describe the evolution of the states. The presence of conjugacy enables us to use a mean-field approximation to update the states given fresh observations. The method is evaluated using the full history of the English Premier League and shown to be competitive, or superior, to weighted likelihood or score-driven time-series-based methods.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press

    Spatiotemporal modelling with dynamic deformation for nonstationary covariance structures

    Fidel Ernesto Castro MoralesMarina Silva Paez
    743-774页
    查看更多>>摘要：In this paper, we present an innovative spatiotemporal model that allows dynamic variation in the spatial correlation structure over time through dynamic deformation. We propose that temporal deformation occurs smoothly relative to that in the original region. To incorporate this idea, we employ state space models to model dynamic deformation. Generalizing this class of models based on spatial deformation was driven by the need to model monthly average temperature data in the southern region of Brazil. The distinctive traits of this region, characterized by plateaus and mountain ranges and close proximity to the Atlantic Ocean, provide notable geographic diversity. This diversity, in addition to different meteorological phenomena over time, may influence the spatial correlation function. The model parameters are estimated via a Bayesian approach, which requires the use of Markov chain Monte Carlo methods to approximate the posterior distributions of parameters. The model is applied to 15 years of monthly average temperature data from the southern region of Brazil. The primary result of this analysis reveals a significant improvement in temperature modelling when the proposed model is used compared with that when versions that employ static deformation are used.
      原文链接:
    • NETL
    • NSTL
    • Oxford Univ Press