A Systematic Review of Spatio-Temporal Statistical Models: Theory, Structure, and Applications
Isabella Habereder, Thomas Kneib, Isao Echizen, Timo Spinde · 2025
Read A Systematic Review of Spatio-Temporal Statistical Models: Theory, Structure, and Applications by Isabella Habereder, Thomas Kneib, Isao Echizen, Timo Spinde (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-00422.txt.
Introduction
Isabella Habereder, Thomas Kneib, Isao Echizen, and Timo Spinde review statistical models for data that vary across both space and time. Such data appear everywhere, but earlier reviews stuck to single fields or model families. Following PRISMA guidelines, the team searched two databases for work from 2021 to 2025 and kept eighty-three publications that met their criteria.
They propose a classification scheme for model structures and map where each structure is used: epidemiology, ecology, public health, economics, and criminology lead the list. Hierarchical models dominate, and most models add components to capture space-time dependence, though preferred shapes differ by field. Research clusters in a few disciplines despite wider relevance, and reproducibility stays limited. The review invites cross-field comparison and more transparent practice.
Catalog record
| CREATOR | Isabella Habereder, Thomas Kneib, Isao Echizen, Timo Spinde |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 91 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-00422
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A Systematic Review of Spatio-Temporal Statistical Models: Theory, Structure, and Applications Isabella Habereder, Thomas Kneib, Isao Echizen, Timo Spinde Abstract Data with spatial-temporal attributes are prevalent across many research fields, and statistical models for analyzing spatio-temporal relationships are widely used. Existing reviews focus either on specific domains or model types, creating a gap in comprehensive, cross-disciplinary overviews. To address this, we conducted a systematic literature review following the PRISMA guidelines, searched two databases for the years 2021-2025, and identified 83 publications that met our criteria. We propose a classification scheme for spatio-temporal model structures and highlight their application in the most common fields: epidemiology, ecology, public health, economics, and criminology. Although tasks vary by domain, many models share similarities. We found that hierarchical models are the most frequently used, and most models incorporate additive components to account for spatial-temporal dependencies. The preferred model structures differ among fields of application. We also observe that research efforts are concentrated in only a few specific disciplines, despite the broader relevance of spatio-temporal data. Furthermore, we notice that reproducibility remains limited. Our review, therefore, not only offers inspiration for comparing model structures in an interdisciplinary manner but also highlights opportunities for greater transparency, accessibility, and cross-domain knowledge transfer. keywords Spatio-Temporal Statistical Models , Systematic Review , PRISMA , , , 1 Introduction Spatio-temporal data 1 1 1 By spatio-temporal data, we refer to data with attributes of both time and space [ 26 ] arise in a wide range of application areas, including epidemiology [ 3 ] , environmental research [ 22 ] , economics [ 28 ] , social sciences [ 90 , 91 ] , and health [ 8 ] . Such data are generated by sensors, digital media platforms, socioeconomic surveys, and environmental monitoring systems, all of which record measurements across space and time [ 26 ] . The diversity of fields in which spatio-temporal data occur gives rise to a broad spectrum of tasks for advanced statistical modeling. Typical applications include modeling the spread of infectious diseases [ 3 , 70 , 98 ] and assessing the health effects of air pollution [ 8 , 63 , 83 ] . We refer to the statistical models used for these purposes as spatio-temporal models. A spatio-temporal statistical model is a mathematical framework that describes and analyzes data that vary across space and time [ 26 ] . The challenge lies in specifying the structure of these models to ensure that relationships across spatial units and time points adequately reflect the underlying phenomenon. Several categories of spatio-temporal models exist, including regression models, point processes, and stochastic compartment models, each encompassing distinct modeling strategies. Researchers make a common distinction between descriptive (in this context, i.e., covariance-based) approaches and dynamic processes based on conditional probability [ 95 , 107 , 109 ] . Despite the widespread use of these methods, comprehensive reviews remain scarce. One foundational work in the field is [ 26 ] , which provides a theoretically rigorous classification of spatio-temporal processes based on data structure and methodology. However, this treatment is highly technical and includes only limited practical examples. Other reviews focus on a single domain (e.g., [ 4 , 17 , 64 , 72 ] ) or on a specific model class, such as point processes (e.g., [ 38 , 81 ] ). Furthermore, we note a discrepancy between the models described in textbooks and theoretical overview works and the actual model structures used in practice. Consequently, there is a lack of up-to-date systematic overviews (see Section 3 ) that integrate perspectives across model types and application areas while also providing a basic classification scheme for the used spatio-temporal model structures. This work aims to present a structured and comparative overview of spatio-temporal model structures and to discuss their applications across disciplines. Such a literature review is overdue, and given the widespread use of spatio-temporal methods, it can benefit various domains by increasing awareness of available techniques and challenges, as well as summarizing the latest state-of-the-art developments. We address the following research questions: (RQ1) What are the most commonly used spatio-temporal model structures in statistical research, and how frequently are they applied across domains? (RQ2) In which domains are spatio-temporal models currently applied, and for what types of problems? (RQ3) What are the key challenges and limitations in the application of spatio-temporal models across different domains? We structure our work into several main sections. First, in Section 2 , we describe the methodology of our systematic literature review. Then, in Section 3 , we embed our work within the existing field of research. We present the results of our review in Section 4 , which forms the central part of the work. We detail the outcomes of the literature search in Section 4.1 . Section 4.2 provides an overview of the spatio-temporal model structures employed. These model structures are then considered in domain-specific contexts in Section 4.3 . Finally, Sections 5 and 6 contain the discussion and conclusion, respectively. All resources for our review are publicly available at https://github.com/Media-Bias-Group/Spatio-Temporal-Statistical-Models-and-Their-Applications . 2 Methodology We conduct a systematic review of published research that applies spatio-temporal statistical methods. The primary contribution of our work is a comprehensive and structured synthesis of both the modeling frameworks and their applications across different domains. Literature reviews are often vulnerable to incomplete coverage and weaknesses in the selection, organization, and presentation of material [ 31 ] , particularly when comprehensive coverage is desired. To minimize these risks, we carefully designed our collection and selection procedures, focusing on consistency and transparency throughout the review process. We adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [ 75 ] , incorporating the following elements: Eligibility criteria (see Section 2.1 ), information sources (see Section 2.2 ), search strategy (see Section 2.2 ), selection process (see Section 2.3 ), data collection process (see Section 2.5 ), data items (see Section 2.5 ), and study risk of bias assessment (see Section 2.6 ). 2.1 Eligibility Criteria Our systematic literature review encompasses peer-reviewed journal articles and conference papers, including only studies published in English. We assume that current studies build upon knowledge accumulated in previous time periods. While many methods in the domain of spatio-temporal statistics originated many years ago (e.g., [ 9 ] ), we aim to review methods used within the last few years, regardless of their initial development date. Therefore, we decided to cover the period from 2021 to 2025. We focus on research on spatio-temporal statistical modeling in applied domains. Theoretical works without application, as well as spatial-only or temporal-only approaches, were not included in the review. We also exclude pure mathematical work without a domain context or any work lacking a modeling focus. Additionally, we do not analyze publications in journals with a Scimago ranking lower than Q1 2 2 2 https://www.scimagojr.com/journalrank.php access 07.08.2025 or from conferences with a core rank lower than A 3 3 3 https://portal.core.edu.au/conf-ranks/ access 07.08.2025 . 2.2 Information Sources and Search Strategy We queried the databases Scopus and Web of Science. Scopus covers a broader range of journals than Web of Science and offers faster citation analysis with greater article coverage [ 31 ] . However, Web of