Spatial Statistics最新文献

{"title":"Geographically informed graph neural networks","authors":"Xuankai Ma ,&nbsp;Zehua Zhang ,&nbsp;Yongze Song","doi":"10.1016/j.spasta.2025.100920","DOIUrl":"10.1016/j.spasta.2025.100920","url":null,"abstract":"<div><div>Graph neural networks (GNNs) have been introduced to spatial statistical tasks due to their mechanisms of simulating spatial interactions and processes among geographical neighbours using graph structures. However, previous methods ignore quantifying differences in attributes among adjacent spatial characteristics. Considering this spatial characteristic by fitting the spatial statistic trinity (SST) framework may help improve models’ accuracy and robustness. Thus, we introduce the geographically informed graph neural network (GIGNN) by considering the additional geospatial feature: closer geographical entities may interact less when spatial disparities are captured. When setting up the model, GIGNN leverages differences of attributes by spatial stratified heterogeneity, quantifies connections between geographical entities, and inherits k-order neighbour attribute aggregation and message-passing mechanisms from GNNs. GIGNN is applied to an urbanization analysis study in the Greater Perth Area, Australia, showing higher accuracy than the existing machine learning models and other GNNs in simulation and prediction accuracy. GIGNN achieved an accuracy of 84.1% for simulation and an accuracy of 81% for prediction. Incorporating spatial characteristics into GNNs enhances simulation and prediction accuracy in geoscientific applications, highlighting the importance of spatially aware models in solving complex problems by capturing geographical data dependencies.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"69 ","pages":"Article 100920"},"PeriodicalIF":2.1,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657301","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Flexible space–time models for extreme data","authors":"Lorenzo Dell’Oro ,&nbsp;Carlo Gaetan","doi":"10.1016/j.spasta.2025.100916","DOIUrl":"10.1016/j.spasta.2025.100916","url":null,"abstract":"<div><div>Extreme value analysis is an essential methodology in the study of rare and extreme events, which hold significant interest in various fields, particularly in the context of environmental sciences. Models that employ the exceedances of values above suitably selected high thresholds possess the advantage of capturing the “sub-asymptotic” dependence of data. This paper presents an extension of spatial random scale mixture models to the spatio-temporal domain. A comprehensive framework for characterizing the dependence structure of extreme events across both dimensions is provided. Indeed, the model is capable of distinguishing between asymptotic dependence and independence, both in space and time, through the use of parametric inference. The high complexity of the likelihood function for the proposed model necessitates a simulation approach based on neural networks for parameter estimation, which leverages summaries of the sub-asymptotic dependence present in the data. The effectiveness of the model in assessing the limiting dependence structure of spatio-temporal processes is demonstrated through both simulation studies and an application to rainfall datasets.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"68 ","pages":"Article 100916"},"PeriodicalIF":2.1,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144631897","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Some peculiar families of correlation functions","authors":"D. Posa","doi":"10.1016/j.spasta.2025.100915","DOIUrl":"10.1016/j.spasta.2025.100915","url":null,"abstract":"<div><div>In this paper a generalization of some families of correlation functions has been proposed; in particular, a generalization of the rational correlation family, as well as a generalization of a subclass of Matérn family have given, together with some relevant properties involving the two classes. Moreover, an extension of the subclass of Matérn family for the two-dimensional and three-dimensional Euclidean spaces has been provided; in addition, the importance of the proposed models for analysing temporal, spatial and, more generally, spatio-temporal data has been underlined, since the same models can be utilized to construct separable as well as non separable correlation functions. It will be shown that these new classes of models are flexible enough to describe both positive and negative correlation structures. On the other hand, with respect to the classical negative correlation models, the proposed families present some features which cannot be found in the same classical negative correlation functions: these relevant properties allow to get new flexible models, which can be helpful for practitioners to accommodate further case studies, as will be shown through some applications.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"69 ","pages":"Article 100915"},"PeriodicalIF":2.1,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657302","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A two-step sampling strategy to improve the prediction accuracy of contamination hotspots and identify hotspot boundaries","authors":"Joonmyoung Kim ,&nbsp;Seonwoo Lee ,&nbsp;Taekseon Ryu ,&nbsp;Jonghyun Na ,&nbsp;Taehyun Yun ,&nbsp;Jeongho Lee ,&nbsp;Hansuk Kim ,&nbsp;Man Jae Kwon ,&nbsp;Ho Young Jo ,&nbsp;Yongsung Joo","doi":"10.1016/j.spasta.2025.100918","DOIUrl":"10.1016/j.spasta.2025.100918","url":null,"abstract":"<div><div>Efficient soil remediation, both economically and environmentally, depends on accurate mapping of contaminant concentrations and boundaries of hotspots (areas with concentrations exceeding a critical threshold) through an effective allocation of limited soil sampling sites. This paper introduces a novel two-step sampling location selection method, referred to as the weighted stepwise spatial sampling (WSSS) method. The WSSS method is specifically designed to provide accurate estimates of contaminant concentrations within hotspots and their boundaries. In the first step, dispersed sampling locations are selected for broad exploration, while in the second step, guided by the digital soil mapping results based on the first-step sampling data, sampling locations are selected to focus on identifying potential hotspots. A simulation study using total petroleum hydrocarbon soil data from South Korea demonstrates the superior accuracy and stability of the WSSS in identifying hotspot boundaries and predicting contaminant concentrations within hotspots, compared to other sampling location selection methods. This performance is achieved through an objective function specifically designed to ensure that the selection of sampling locations in the second step is robust to potential inaccuracies or uncertainties in the initial predictions.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"69 ","pages":"Article 100918"},"PeriodicalIF":2.1,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial robust fuzzy clustering of mixed data with electoral study","authors":"Domenico Cangemi ,&nbsp;Pierpaolo D’Urso ,&nbsp;Livia De Giovanni ,&nbsp;Lorenzo Federico ,&nbsp;Vincenzina Vitale","doi":"10.1016/j.spasta.2025.100914","DOIUrl":"10.1016/j.spasta.2025.100914","url":null,"abstract":"<div><div>A robust fuzzy clustering model for data with mixed features and spatial constraints is proposed to analyze the turnout and the preferences of the voters at the provincial level in the European elections. The 2024 European elections in Italy were held in June to elect the 76 members of the European Parliament due to Italy. The clustering model accommodates various types of variables or attributes by integrating dissimilarity measures for each one through a weighting approach. This method produces a composite distance (or dissimilarity) metric that captures multiple attribute types. The weights are determined objectively during the optimization process and indicate the importance of each attribute type. The model also incorporates robustness via the introduction of a Noise cluster, and accounts for a spatial component. The application shows consistency of the results both at the level of units’ attributes and at a spatial level.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"69 ","pages":"Article 100914"},"PeriodicalIF":2.1,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A marked sequential point process for disease surveillance: Modeling and optimization","authors":"François d’Alayer,&nbsp;Edith Gabriel,&nbsp;Samuel Soubeyrand","doi":"10.1016/j.spasta.2025.100913","DOIUrl":"10.1016/j.spasta.2025.100913","url":null,"abstract":"<div><div>Plant disease surveillance is essential for the management of disease outbreaks that pose significant threats to agricultural sustainability. In this study, we present a novel sequential point process model designed for disease surveillance. The model incorporates self-interaction mechanisms to account for the influence of the process’ history. To analyze the dynamics of the model, we propose new sequential summary statistics that extend traditional point process methods to scenarios where sequential interactions are critical. This model serves a dual purpose: it is employed both to propose novel and efficient sampling designs, and to characterize existing sampling schemes, implemented in real-world situations, through parameter inference.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"68 ","pages":"Article 100913"},"PeriodicalIF":2.1,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144562979","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Attribute based spatial segmentation for optimising POI placement","authors":"M. de Klerk,&nbsp;I. Fabris-Rotelli","doi":"10.1016/j.spasta.2025.100911","DOIUrl":"10.1016/j.spasta.2025.100911","url":null,"abstract":"<div><div>Effective spatial planning and resource optimisation require precise demarcation of potential spatial accessible areas and optimal placement of points of interest (POIs). Our approach introduces a novel attribute based spatial segmentation methodology that utilises an iterative clustering approach to create unique macro-regions, each associated with key structural and attribute specific properties. By integrating a probabilistic attribute based structure with k-means clustering, we adaptively segment spatial regions to balance area based attributes and topological characteristics. The full geographical network is segmented into attribute based macro-regions for all spatially accessible and spatially disjoint regions. Attribute based spatial segmentation offers insights into why certain areas may be spatially disjoint and if it is identified as potential spatially accessible areas to determine which POIs can be placed to maximise accessibility. This approach transforms city planning and resource allocation by aligning POI placement with regional needs and characteristics.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"68 ","pages":"Article 100911"},"PeriodicalIF":2.1,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144480059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transfer learning for high dimensional spatial autoregressive model","authors":"Yunquan Song,&nbsp;Xuan Chen,&nbsp;Rui Yang,&nbsp;Yijun Li","doi":"10.1016/j.spasta.2025.100908","DOIUrl":"10.1016/j.spasta.2025.100908","url":null,"abstract":"<div><div>Transfer learning is a learning process that applies models learned in old domains to new domains by utilizing similarities between data, tasks, or models. At present, transfer learning has been widely applied, such as natural language processing, recommendation systems, drug analysis, etc. Research in statistical models mostly focuses on classic linear models such as classification and regression. It is still unclear how transfer learning affects spatial data. Spatial data is an important type of data and has been a hot research topic in statistics and econometrics in recent years. However, in reality, its collection and labeling are expensive and labor-intensive, and there may not be enough data to train a robust model. Therefore, this article considers using auxiliary sample sets that are different from the target dataset but have some similarity to help us estimate and predict the target model, and specifies criteria for determining similarity. We propose transfer learning algorithms based on spatial autoregressive models, which can transfer knowledge from auxiliary datasets to target models of interest to us. Its performance has been demonstrated in numerical simulations and real housing price datasets.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"68 ","pages":"Article 100908"},"PeriodicalIF":2.1,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144289162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characteristics of some isotropic covariance models with negative values","authors":"De Iaco S.,&nbsp;Posa D.","doi":"10.1016/j.spasta.2025.100905","DOIUrl":"10.1016/j.spasta.2025.100905","url":null,"abstract":"<div><div>In the literature, most of the classical covariance models characterised by negative values were derived by utilising the Bessel functions, on the other hand, recently, other classes of models with negative correlation were obtained through the difference between two covariance functions. However, although for the former, the analytic features, such as their absolute minimum values, were completely explored, for the latter these aspects have to be still investigated. In this paper, starting from the admissibility conditions and the general characteristics of three wide families of isotropic covariance models, based on the difference of Gaussian, exponential and rational models, their absolute minimum, as a function of the dimension of the Euclidean space in which they are defined, is provided. Consequently, the minimum values for the most common Euclidean dimensional spaces are given as special cases. These results fill the theoretical gap related to the analysed classes of correlation models with negative values and then can support their use. A simulation study and an application to a real data set are also presented to assess performance in terms of prediction accuracy.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"68 ","pages":"Article 100905"},"PeriodicalIF":2.1,"publicationDate":"2025-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144479932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatiotemporal mapping and analysis of atypical COVID-19 outbreaks in Shijiazhuang City (China) using the synthetic SEIR-BME approach","authors":"Zekun Gao ,&nbsp;Yutong Jiang ,&nbsp;Junjie Yin ,&nbsp;Jiaping Wu ,&nbsp;Maria-Stephania Christakos ,&nbsp;George Christakos ,&nbsp;Junyu He","doi":"10.1016/j.spasta.2025.100910","DOIUrl":"10.1016/j.spasta.2025.100910","url":null,"abstract":"<div><div>Shijiazhuang City (Hebei Province, China) experienced two COVID-19 outbreaks: January 2021 and November 2022. Differences in the prevention and control measures implemented during the two outbreaks led to significantly distinct epidemic evolutions. During the first outbreak, these measures were implemented throughout the epidemic duration. During the second outbreak, attention was paid only at the initial epidemic stage, followed by a laissez-faire management that led to a rapid epidemic development, and only then control measures were re-implemented. In the present work, epidemic-related data during the two outbreaks and relevant risk area data during the atypical November 2022 outbreak were collected from Nation-, Hebei Province-, and Shijiazhuang City-level Health Commission sources. The study of the outbreaks involved a preliminary long time-series analysis followed by a novel synthesis of Susceptible-Exposed-Infected-Removed (SEIR) modeling with Bayesian Maximum Entropy (BME) mapping of the spatiotemporal COVID-19 spread during the November 2022 outbreak (a severe data deficiency occurred this month due to normalized management). An important advantage of the proposed SEIR-BME synthesis is that it compensated for the individual shortcomings of its components: Using SEIR we constructed transmission models of the outbreaks, while BME effectively filled in the missing data during November 2022 and subsequently generated accurate spatiotemporal disease risk maps. Our results confirmed the powerful transmission capability of COVID-19 and the considerable prevention and control progress made by the authorities from January 2021 to November 2022. We also found that during the exponential growth period of the epidemic, the COVID-19 variation results of this work closely followed the empirical COVID-19 law of <span><span>He et al. (2020)</span></span>. Lastly, our analysis provided data support for subsequent studies of the COVID-19 spread, and suggested optimal infectious disease prevention and control measures. It is hoped that the present work would laid the methodological foundations for future developments in spatiotemporal infectious disease modeling and mapping.</div></div>","PeriodicalId":48771,"journal":{"name":"Spatial Statistics","volume":"68 ","pages":"Article 100910"},"PeriodicalIF":2.1,"publicationDate":"2025-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}