BEGIN:VCALENDAR VERSION:2.0 PRODID:-//pretalx//talks.osgeo.org//foss4g-2026//talk//BHEB3E BEGIN:VTIMEZONE TZID:JST BEGIN:STANDARD DTSTART:20000101T000000 RRULE:FREQ=YEARLY;BYMONTH=1 TZNAME:JST TZOFFSETFROM:+0900 TZOFFSETTO:+0900 END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:pretalx-foss4g-2026-BHEB3E@talks.osgeo.org DTSTART;TZID=JST:20260902T163000 DTEND;TZID=JST:20260902T170000 DESCRIPTION:Open geospatial data have become increasingly available in rece nt years\, enabling researchers and practitioners to analyze complex spati al phenomena at unprecedented spatial and temporal resolutions. Government statistics\, environmental monitoring networks\, and socio-economic indic ators are now widely distributed through open data platforms and can be in tegrated with geographic information systems (GIS). However\, many spatial datasets contain multiple interrelated variables that evolve simultaneous ly across both space and time. Examples include regional economic indicato rs\, demographic statistics\, environmental measurements\, and infrastruct ure activity. Understanding the interactions among these variables is esse ntial for studying spatial systems\, yet most existing spatial modeling ap proaches treat each variable separately.\n\nTraditional spatial regression models such as the spatial autoregressive (SAR) model and the spatial err or model (SEM) are widely used in spatial econometrics and spatial statist ics. These models capture spatial dependence through spatial weight matric es representing relationships among geographic units such as administrativ e regions or grid cells. While these methods have proven useful for analyz ing spatial spillovers and diffusion processes\, they are typically formul ated for a single response variable. When multiple spatial variables are a nalyzed separately\, potential interactions among them cannot be represent ed explicitly. As a result\, the joint dynamics of spatial systems may rem ain partially unexplained.\n\nThis presentation introduces a multivariate spatio-temporal regression framework designed to analyze multiple spatial variables simultaneously. The proposed model extends classical spatial reg ression models to a multivariate setting by representing dependent variabl es as vectors that evolve across both geographic space and time. In this f ramework\, spatial dependence is represented through spatial weight matric es commonly used in GIS-based spatial analysis\, while temporal dependence is incorporated through autoregressive structures. The resulting model ca ptures three important types of interactions: spatial spillovers across ne ighboring regions\, temporal persistence within each variable\, and cross- variable interactions among multiple spatial indicators.\n\nThe proposed m odeling framework is expressed in a compact matrix form and estimated usin g maximum likelihood methods. The likelihood-based estimation approach ena bles efficient parameter estimation while maintaining statistical interpre tability. Under suitable conditions\, the parameters of the model are iden tifiable\, meaning that spatial effects and cross-variable interactions ca n be uniquely recovered from observed spatial data. Because the model stru cture generalizes classical spatial econometric specifications\, it natura lly nests widely used models such as multivariate spatial autoregressive a nd spatial error models.\n\nTo evaluate the statistical performance of the proposed framework\, Monte Carlo simulation experiments were conducted un der a variety of spatial and temporal dependence scenarios. The simulation s demonstrate that the estimation procedure can accurately recover true mo del parameters even when complex cross-variable spatial interactions are p resent. These results indicate that the multivariate formulation provides a robust statistical tool for modeling multidimensional spatial dependence .\n\nThe practical applicability of the approach is demonstrated using reg ional data from Japan. The empirical analysis focuses on the joint spatial dynamics of prefectural fertility rates and regional gross domestic produ ct (GDP). These variables are closely related through demographic and econ omic processes and are likely to influence one another across neighboring regions. Using GIS-based regional datasets and spatial weight matrices rep resenting prefectural adjacency relationships\, the multivariate spatio-te mporal model is estimated to capture both spatial spillovers and cross-var iable interactions. Model comparison based on the Akaike Information Crite rion indicates that the multivariate specification provides a better fit t o the data than conventional univariate spatial regression models. The res ults suggest that explicitly modeling multidimensional spatial dependence improves both statistical performance and interpretability of regional pro cesses.\n\nBeyond this specific example\, the proposed framework has broad implications for spatial data science and geospatial analytics. Many open geospatial datasets—including environmental observations\, urban indica tors\, transportation data\, and socio-economic statistics—contain multi ple variables that interact across both space and time. The multivariate s patio-temporal regression framework provides a statistical methodology for analyzing such datasets while accounting for spatial dependence structure s commonly used in GIS analysis. Because the model relies on spatial weigh t matrices and likelihood-based estimation\, it can be integrated with exi sting geospatial workflows and open-source statistical environments. It is also potentially useful for policy evaluation\, regional forecasting\, di saster recovery assessment\, and evidence-based planning using linked spat ial indicators.\n\nFor the FOSS4G community\, this work highlights the imp ortance of combining statistical modeling with open geospatial infrastruct ures. While GIS platforms provide powerful tools for visualizing and manag ing spatial data\, rigorous statistical models are essential for understan ding spatial interactions and making reliable inferences from complex data sets. The proposed modeling approach contributes to this integration by of fering a statistically grounded framework for analyzing multidimensional s patial processes using GIS-based regional data.\n\nFuture work will focus on implementing the proposed modeling framework in open-source statistical and geospatial software environments\, enabling researchers and practitio ners to apply multivariate spatial modeling techniques to a wide range of open geospatial datasets. By bridging statistical methodology and open geo spatial data analysis\, the approach aims to support more comprehensive mo deling of complex spatial systems in both research and applied domains. DTSTAMP:20260717T225734Z LOCATION:Cosmos1 SUMMARY:Multivariate Spatio-Temporal Modeling for Regional GIS Data: A Stat istical Framework for Analyzing Multidimensional Spatial Interactions - Sa eko Ohta URL:https://talks.osgeo.org/foss4g-2026/talk/BHEB3E/ END:VEVENT END:VCALENDAR