Pub Date : 2025-12-22DOI: 10.1146/annurev-statistics-042324-040052
Xiuyuan Cheng, Zheng Dong, Yao Xie
Spatiotemporal point processes model discrete events distributed in space and time, with applications in criminology, seismology, epidemiology, and social networks. Classical models rely on parametric kernels, limiting their ability to capture heterogeneous, nonstationary dynamics. Recent advances integrate deep neural architectures, either by modeling the conditional intensity directly or by learning flexible, data-driven influence kernels. This article reviews the deep influence kernel approach, which balances statistical interpretability by retaining explicit kernels to capture event propagation, with expressive power from neural architectures. We outline key components, including functional basis decomposition, graph neural networks for encoding spatial or network structures, and both likelihood-based and likelihood-free estimation methods, while addressing scalability for large data. We also highlight theoretical results on kernel identifiability. Applications in crime analysis, earthquake aftershock prediction, and sepsis modeling demonstrate the framework's effectiveness. We conclude with promising directions for developing explainable and scalable deep kernel point processes.
{"title":"Deep Spatiotemporal Point Processes: Advances and New Directions","authors":"Xiuyuan Cheng, Zheng Dong, Yao Xie","doi":"10.1146/annurev-statistics-042324-040052","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042324-040052","url":null,"abstract":"Spatiotemporal point processes model discrete events distributed in space and time, with applications in criminology, seismology, epidemiology, and social networks. Classical models rely on parametric kernels, limiting their ability to capture heterogeneous, nonstationary dynamics. Recent advances integrate deep neural architectures, either by modeling the conditional intensity directly or by learning flexible, data-driven influence kernels. This article reviews the deep influence kernel approach, which balances statistical interpretability by retaining explicit kernels to capture event propagation, with expressive power from neural architectures. We outline key components, including functional basis decomposition, graph neural networks for encoding spatial or network structures, and both likelihood-based and likelihood-free estimation methods, while addressing scalability for large data. We also highlight theoretical results on kernel identifiability. Applications in crime analysis, earthquake aftershock prediction, and sepsis modeling demonstrate the framework's effectiveness. We conclude with promising directions for developing explainable and scalable deep kernel point processes.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"84 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145908310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-03DOI: 10.1146/annurev-statistics-112723-034107
Eric C. Chi, Aaron J. Molstad, Zheming Gao, Jocelyn T. Chi
This article reviews a clustering method based on solving a convex optimization problem. Despite the plethora of existing clustering methods, convex clustering has several uncommon features that distinguish it from its predecessors. The optimization problem is free of spurious local minima, and its unique global minimizer is stable with respect to all its inputs, including the data, a tuning parameter, and weight hyperparameters. Its single tuning parameter controls the number of clusters and can be chosen using standard techniques from penalized regression. We give intuition into the behavior of and theory for convex clustering, as well as practical guidance. We highlight important algorithms and discuss how their computational costs scale with the problem size. Finally, we highlight the breadth of its uses and flexibility to be combined and integrated with other inferential methods.
{"title":"The Why and How of Convex Clustering","authors":"Eric C. Chi, Aaron J. Molstad, Zheming Gao, Jocelyn T. Chi","doi":"10.1146/annurev-statistics-112723-034107","DOIUrl":"https://doi.org/10.1146/annurev-statistics-112723-034107","url":null,"abstract":"This article reviews a clustering method based on solving a convex optimization problem. Despite the plethora of existing clustering methods, convex clustering has several uncommon features that distinguish it from its predecessors. The optimization problem is free of spurious local minima, and its unique global minimizer is stable with respect to all its inputs, including the data, a tuning parameter, and weight hyperparameters. Its single tuning parameter controls the number of clusters and can be chosen using standard techniques from penalized regression. We give intuition into the behavior of and theory for convex clustering, as well as practical guidance. We highlight important algorithms and discuss how their computational costs scale with the problem size. Finally, we highlight the breadth of its uses and flexibility to be combined and integrated with other inferential methods.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"10 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145658281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-26DOI: 10.1146/annurev-statistics-042324-061403
Jay S. Kaufman
Measurement and analysis of racial and ethnic health disparities are vital functions of government and academia in diverse societies, but the statistical methods for accomplishing this work are underdeveloped. Issues of measurement, aggregation, adjustment, choice of scale, internal validity, and generalizability are all paramount. Measurement of race and ethnicity is complicated by the fact that, as identities that form through historical and political processes, they are not stable over time and place, nor are they objectively verifiable. Similarly, it is impossible to specify an optimal adjustment set, because adjustments are functions of ethical judgments, not statistical criteria. Additional complications arise when decomposing disparities in relation to measured pathways, as well as in the modeling of multiple intersectional strata. The ethical considerations in model selection imply that measurement and modeling of health disparities can never be a purely statistical activity, but instead must be conducted in relation to a theory of justice.
{"title":"Statistical Aspects of Racial and Ethnic Health Disparities","authors":"Jay S. Kaufman","doi":"10.1146/annurev-statistics-042324-061403","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042324-061403","url":null,"abstract":"Measurement and analysis of racial and ethnic health disparities are vital functions of government and academia in diverse societies, but the statistical methods for accomplishing this work are underdeveloped. Issues of measurement, aggregation, adjustment, choice of scale, internal validity, and generalizability are all paramount. Measurement of race and ethnicity is complicated by the fact that, as identities that form through historical and political processes, they are not stable over time and place, nor are they objectively verifiable. Similarly, it is impossible to specify an optimal adjustment set, because adjustments are functions of ethical judgments, not statistical criteria. Additional complications arise when decomposing disparities in relation to measured pathways, as well as in the modeling of multiple intersectional strata. The ethical considerations in model selection imply that measurement and modeling of health disparities can never be a purely statistical activity, but instead must be conducted in relation to a theory of justice.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"118 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145609859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-21DOI: 10.1146/annurev-statistics-042424-070908
Unique Subedi, Ambuj Tewari
Operator learning has emerged as a powerful tool in scientific computing for approximating mappings between infinite-dimensional function spaces. A primary application of operator learning is the development of surrogate models for the solution operators of partial differential equations (PDEs). These methods can also be used to develop black-box simulators to model system behavior from experimental data, even without a known mathematical model. In this article, we begin by formalizing operator learning as a function-to-function regression problem and review some recent developments in the field. We also discuss PDE-specific operator learning, outlining strategies for incorporating physical and mathematical constraints into architecture design and training processes. Finally, we end by highlighting key future directions such as active data collection and the development of rigorous uncertainty quantification frameworks.
{"title":"Operator Learning: A Statistical Perspective","authors":"Unique Subedi, Ambuj Tewari","doi":"10.1146/annurev-statistics-042424-070908","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042424-070908","url":null,"abstract":"Operator learning has emerged as a powerful tool in scientific computing for approximating mappings between infinite-dimensional function spaces. A primary application of operator learning is the development of surrogate models for the solution operators of partial differential equations (PDEs). These methods can also be used to develop black-box simulators to model system behavior from experimental data, even without a known mathematical model. In this article, we begin by formalizing operator learning as a function-to-function regression problem and review some recent developments in the field. We also discuss PDE-specific operator learning, outlining strategies for incorporating physical and mathematical constraints into architecture design and training processes. Finally, we end by highlighting key future directions such as active data collection and the development of rigorous uncertainty quantification frameworks.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"29 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145567648","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-18DOI: 10.1146/annurev-statistics-042324-060005
Deependra K. Thapa, Erik S. Parker, Mounika Kandukuri, Xi (Rita) Wang, Thirupathi R. Mokalla, Olivia C. Robertson, Wasiuddin Najam, Andrew E. Teschendorff, Andrew W. Brown, John R. Speakman, Yisheng Peng, Bernard S. Gorman, Heping Zhang, Luis-Enrique Becerra-Garcia, Colby J. Vorland, David B. Allison
Aging research relies on varied statistical methods, and applying these methods appropriately is important for scientific rigor. However, proper use of these statistical techniques is a challenge. We discuss two categories of statistical methods in aging research: ( a ) emerging methods requiring further validation, including techniques to examine compression of morbidity, maximum lifespan, immortal time bias, molecular aging clocks, and treatment response heterogeneity, and ( b ) classic and existing methods needing reconsideration and improvement, such as stepwise regression, generalized linear models, methods for accounting for clustering and nesting effects, methods for testing for group differences, methods for mediation and moderation analyses, and nonlinear models. For each method, we review its relevance to aging research, highlight statistical issues, and suggest improvements or alternatives with examples from aging research. We urge researchers to refine traditional approaches and embrace emerging methods tailored to the unique challenges of aging research. This review will help researchers identify and apply sound statistical methods, thereby improving statistical rigor in aging research.
{"title":"Statistical Methods in Aging Research: Improving Current Practices and Embracing Emerging Approaches","authors":"Deependra K. Thapa, Erik S. Parker, Mounika Kandukuri, Xi (Rita) Wang, Thirupathi R. Mokalla, Olivia C. Robertson, Wasiuddin Najam, Andrew E. Teschendorff, Andrew W. Brown, John R. Speakman, Yisheng Peng, Bernard S. Gorman, Heping Zhang, Luis-Enrique Becerra-Garcia, Colby J. Vorland, David B. Allison","doi":"10.1146/annurev-statistics-042324-060005","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042324-060005","url":null,"abstract":"Aging research relies on varied statistical methods, and applying these methods appropriately is important for scientific rigor. However, proper use of these statistical techniques is a challenge. We discuss two categories of statistical methods in aging research: ( <jats:italic>a</jats:italic> ) emerging methods requiring further validation, including techniques to examine compression of morbidity, maximum lifespan, immortal time bias, molecular aging clocks, and treatment response heterogeneity, and ( <jats:italic>b</jats:italic> ) classic and existing methods needing reconsideration and improvement, such as stepwise regression, generalized linear models, methods for accounting for clustering and nesting effects, methods for testing for group differences, methods for mediation and moderation analyses, and nonlinear models. For each method, we review its relevance to aging research, highlight statistical issues, and suggest improvements or alternatives with examples from aging research. We urge researchers to refine traditional approaches and embrace emerging methods tailored to the unique challenges of aging research. This review will help researchers identify and apply sound statistical methods, thereby improving statistical rigor in aging research.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"101 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145545454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-17DOI: 10.1146/annurev-statistics-042424-052503
Jiguo Cao, Sidi Wu, Muye Nanshan, Haolun Shi, Liangliang Wang
Functional data analysis (FDA) is a rapidly growing field in modern statistics that provides powerful tools for analyzing data observed as curves, surfaces, or more general functions. Unlike traditional multivariate methods, FDA explicitly accounts for the smooth and continuous nature of functional data, enabling more accurate modeling and interpretation. Traditional FDA methods, such as functional principal component analysis, functional regression, and functional classification, rely on linear assumptions and basis function expansions, which can limit their effectiveness when applied to nonlinear, high-dimensional, or irregularly sampled data. Recent advances in neural networks provide promising alternatives to these traditional approaches. Deep learning methods offer several key advantages: They naturally capture nonlinear relationships, scale to high-dimensional data without explicit dimension reduction, learn task-specific representations directly from raw observations, and handle sparse or irregular sampling without requiring imputation. This article reviews recent methodological developments in FDA, with a focus on the integration of deep learning techniques. Through this comparative review, we highlight the strengths and limitations of classical and modern approaches, providing practical guidance and future directions.
{"title":"Statistical Learning for Functional Data","authors":"Jiguo Cao, Sidi Wu, Muye Nanshan, Haolun Shi, Liangliang Wang","doi":"10.1146/annurev-statistics-042424-052503","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042424-052503","url":null,"abstract":"Functional data analysis (FDA) is a rapidly growing field in modern statistics that provides powerful tools for analyzing data observed as curves, surfaces, or more general functions. Unlike traditional multivariate methods, FDA explicitly accounts for the smooth and continuous nature of functional data, enabling more accurate modeling and interpretation. Traditional FDA methods, such as functional principal component analysis, functional regression, and functional classification, rely on linear assumptions and basis function expansions, which can limit their effectiveness when applied to nonlinear, high-dimensional, or irregularly sampled data. Recent advances in neural networks provide promising alternatives to these traditional approaches. Deep learning methods offer several key advantages: They naturally capture nonlinear relationships, scale to high-dimensional data without explicit dimension reduction, learn task-specific representations directly from raw observations, and handle sparse or irregular sampling without requiring imputation. This article reviews recent methodological developments in FDA, with a focus on the integration of deep learning techniques. Through this comparative review, we highlight the strengths and limitations of classical and modern approaches, providing practical guidance and future directions.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"155 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145536105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-14DOI: 10.1146/annurev-statistics-042424-050626
Kartik Waghmare, Johanna Ziegel
Proper scoring rules have been a subject of growing interest in recent years, not only as tools for evaluation of probabilistic forecasts but also as methods for estimating probability distributions. In this article, we review the mathematical foundations of proper scoring rules, including general characterization results and important families of scoring rules. We discuss their role in statistics and machine learning for estimation and forecast evaluation. Furthermore, we comment on interesting developments of their usage in applications.
{"title":"Proper Scoring Rules for Estimation and Forecast Evaluation","authors":"Kartik Waghmare, Johanna Ziegel","doi":"10.1146/annurev-statistics-042424-050626","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042424-050626","url":null,"abstract":"Proper scoring rules have been a subject of growing interest in recent years, not only as tools for evaluation of probabilistic forecasts but also as methods for estimating probability distributions. In this article, we review the mathematical foundations of proper scoring rules, including general characterization results and important families of scoring rules. We discuss their role in statistics and machine learning for estimation and forecast evaluation. Furthermore, we comment on interesting developments of their usage in applications.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"26 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145509527","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-07DOI: 10.1146/annurev-statistics-042324-012947
J. López-Fidalgo, W.K. Wong
Correlated data occur naturally and frequently in small and big data, and methods for analyzing correlated data have seen great advances in recent decades. Attention to design issues typically lags behind that paid to estimation issues, and it is also true that construction of optimal designs for models with correlated data lags that for models with uncorrelated data. A key problem in constructing optimal designs for models with correlated observations is that more technical assumptions are needed than when models have uncorrelated errors. In the former case, approximations to the information matrix are needed, and there are also no general and effective algorithms for finding various types of optimal designs. In addition, there are no tools to confirm optimality of a design. This article first gives a short review of optimal designs for linear models, before we focus on a review of finding optimal designs for models with correlated data. We discuss various approaches and their difficulties in a few selected areas. Along the way, we provide examples and recommend use of nature-inspired metaheuristic algorithms to find all kinds of optimal designs for any criterion or model with or without correlated data.
{"title":"Optimal Designs for Correlated Data","authors":"J. López-Fidalgo, W.K. Wong","doi":"10.1146/annurev-statistics-042324-012947","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042324-012947","url":null,"abstract":"Correlated data occur naturally and frequently in small and big data, and methods for analyzing correlated data have seen great advances in recent decades. Attention to design issues typically lags behind that paid to estimation issues, and it is also true that construction of optimal designs for models with correlated data lags that for models with uncorrelated data. A key problem in constructing optimal designs for models with correlated observations is that more technical assumptions are needed than when models have uncorrelated errors. In the former case, approximations to the information matrix are needed, and there are also no general and effective algorithms for finding various types of optimal designs. In addition, there are no tools to confirm optimality of a design. This article first gives a short review of optimal designs for linear models, before we focus on a review of finding optimal designs for models with correlated data. We discuss various approaches and their difficulties in a few selected areas. Along the way, we provide examples and recommend use of nature-inspired metaheuristic algorithms to find all kinds of optimal designs for any criterion or model with or without correlated data.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"21 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145472687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-07DOI: 10.1146/annurev-statistics-042424-110756
Aaron L. Sarvet, Mats J. Stensrud
The natural treatment value (NTV) is the value a treatment takes when it is not altered by an intervention. This observable random variable is foundational to statistical causal inference. On the one hand, identification hinges on our substantive knowledge about the NTV. On the other hand, the NTV is a defining feature of canonical estimands, like the average treatment effect in the treated, local average treatment effects, and natural effects in mediation analysis. In this article, we argue why an explicit and formal consideration of the NTV is important in statistics and related fields, and describe its role in guiding statistical analysis, formulating identification conditions, falsifying assumptions, and relating different estimands. We also review a growing literature studying estimands explicitly defined by the NTV. This allows us to highlight a subtle, often-overlooked identification issue that challenges the study of dynamic regimes that depend on the NTV. Finally, we illustrate how NTV parameters are often motivated by pragmatic concerns, and we consider the practical relevance of some of these estimands.
{"title":"The Natural Value of Treatment and Its Importance for Causal Inference","authors":"Aaron L. Sarvet, Mats J. Stensrud","doi":"10.1146/annurev-statistics-042424-110756","DOIUrl":"https://doi.org/10.1146/annurev-statistics-042424-110756","url":null,"abstract":"The natural treatment value (NTV) is the value a treatment takes when it is not altered by an intervention. This observable random variable is foundational to statistical causal inference. On the one hand, identification hinges on our substantive knowledge about the NTV. On the other hand, the NTV is a defining feature of canonical estimands, like the average treatment effect in the treated, local average treatment effects, and natural effects in mediation analysis. In this article, we argue why an explicit and formal consideration of the NTV is important in statistics and related fields, and describe its role in guiding statistical analysis, formulating identification conditions, falsifying assumptions, and relating different estimands. We also review a growing literature studying estimands explicitly defined by the NTV. This allows us to highlight a subtle, often-overlooked identification issue that challenges the study of dynamic regimes that depend on the NTV. Finally, we illustrate how NTV parameters are often motivated by pragmatic concerns, and we consider the practical relevance of some of these estimands.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"52 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145472685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-06DOI: 10.1146/annurev-statistics-112723-034603
Vianey Leos-Barajas, Ignacio Alvarez-Castro, Juan M. Morales
Advances in technology are paving the way for researchers to remotely track wild animals and collect massive, high-resolution animal movement data sets with temporal and/or spatial structure. However, the rate at which data are becoming available is outpacing the development of statistical methodology that can adequately analyze them. In this article, we cover the most widely used modeling approaches for the analysis of animal movement data and various extensions that have been proposed for each modeling framework, as well as challenges that remain. There are several newer statistical challenges that researchers have tried to tackle in recent years, such as modeling data streams collected at vastly different temporal resolutions from multiple devices to study animal behavior and incorporating physiological processes as drivers of animal movement. We conclude with additional statistical challenges and opportunities that remain to advance the study of animal movement.
{"title":"Statistics for Animal Tracking Data","authors":"Vianey Leos-Barajas, Ignacio Alvarez-Castro, Juan M. Morales","doi":"10.1146/annurev-statistics-112723-034603","DOIUrl":"https://doi.org/10.1146/annurev-statistics-112723-034603","url":null,"abstract":"Advances in technology are paving the way for researchers to remotely track wild animals and collect massive, high-resolution animal movement data sets with temporal and/or spatial structure. However, the rate at which data are becoming available is outpacing the development of statistical methodology that can adequately analyze them. In this article, we cover the most widely used modeling approaches for the analysis of animal movement data and various extensions that have been proposed for each modeling framework, as well as challenges that remain. There are several newer statistical challenges that researchers have tried to tackle in recent years, such as modeling data streams collected at vastly different temporal resolutions from multiple devices to study animal behavior and incorporating physiological processes as drivers of animal movement. We conclude with additional statistical challenges and opportunities that remain to advance the study of animal movement.","PeriodicalId":48855,"journal":{"name":"Annual Review of Statistics and Its Application","volume":"74 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2025-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145455262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：