Jorge Mateu | |
---|---|
![]() | |
Born | Spain | October 11, 1969
Nationality | Spanish |
Occupation(s) | Mathematician, author and academic |
Family | Nadia, Victor, Adrian and Nuria (sons and daughters) |
Academic background | |
Education | Undergraduate., Mathematics and Statistics M.Sc. Mathematics Ph.D. Mathematics |
Alma mater | University of Valencia |
Academic work | |
Institutions | Jaume I University |
Website | "Web site Jorge Mateu". |
Jorge Mateu is a Spanish mathematician, author, and academic. He is a professor of statistics within the Department of Mathematics at University Jaume I of Castellon [1] and director of the Unit Eurocop for Data Science in criminal activities in the same department. [2]
Mateu's research is centered on data science, geostatistics, and stochastic processes, with a particular emphasis on spatio-temporal point processes. [3] He led the 'Mathematical-statistical modelling of space-time data and data mining' group at Universitat Jaume I to develop spatio-temporal statistical techniques used for modelling across fields of public safety, environmental management, and criminology. [4] He is co-editor of books, including Spatial Statistics Through Applications (2002), Case Studies in Spatial Point Process Modeling (2005), Spatio-temporal Design. Advances in Efficient Data Acquisition (2012), Spatial and Spatio-Temporal Geostatistical Modeling and Kriging (2015), or Geostatistical Functional Data Analysis (2021). He has also received the Social Council Award from UJI and has been noted as a World Class Professor by an Indonesian ministry. [5]
Mateu is a Fellow of the Royal Statistical Society and Wessex Institute in Great Britain and a member of The International Statistical Institute [6] and the Bernoulli Society for Mathematical Statistics and Probability. He served as a guest editor for special issues in the Journal of Geophysical Research , and Environmetrics, as the editor-in-chief for the Journal of Agricultural, Biological, and Environmental Statistics [7] as well as an associate editor for Stochastic Environmental Research and Risk Assessment, [8] Spatial Statistics , [9] Environmetrics, [10] and International Statistical Review. [11]
Mateu earned his undergraduate degree in mathematics and statistics from the University of Valencia in 1987, followed by a master's degree in 1995. He graduated with a Ph.D. from the Department of Mathematics at University of Valencia (UV) in 1998. [12]
Mateu began his academic career as an assistant professor of statistics in the Department of Mathematics at Jaume I University in 1992 [13] where he served as an associate professor from 2000 to 2007. In 2007, he assumed the position of Full Professor of Statistics at UJI. [14]
In 2011, he held the position of secretary for the International Environmetrics Society's board of directors [15] and became a co-director of the Erasmus Mundus Master in Geospatial Technologies. [16] Additionally, he served as President of the Board of Editors for METMA Workshops [17] Since 2014, he has been serving as the director of the Unit Eurocop: Statistical Modeling of Crime Data at Jaume I University. [2]
Mateu focuses his research on the intersection of geostatistics, spatial data, stochastic processes, computational sciences, and natural sciences, with a particular emphasis on data science. He has analysed crime data and public health projects by employing a combination of statistical and machine-learning methods. [18] He served as a joint principal investigator for GEO-C. [19] He was worked on the projects (a) Statistical analysis of complex dependencies in space-time stochastic processes. Networks, functional marks and SPDE-based intensities. Ministry of Science and bInnovation (PID2022-141555OB-I00), 2023-2026, and (b) Spatio-temporal stochastic processes over networks and trajectories. Parametric models and functional marks. Generalitat Valenciana (CIAICO/2022/191), 2023-2025.
Mateu's research on data science has included a range of topics such as filament delineation, model selection, and stochastic processes. In his research on the automatic delineation of filaments obtained from redshift catalogs, he applied a marked point process, to gain insights into the cosmic filament structure. [20] Together with a number of coauthors, he extended Gneiting's work to develop new spatio-temporal covariance models, resulting in novel classes of stationary nonseparable functions. [21] In addition, his research of space-time covariance function estimation introduced two methods based on the concept of composite likelihood which were designed to strike a balance between computational complexity and statistical efficiency. [22] Furthermore, while addressing the challenge of model selection, he discussed the limitations of traditional models like Bayesian Information Criterion and proposed a practical extension aimed at handling model selection issues effectively. [23] In 2018, during his research on the use of administrative data, he identified challenges related to statistical analyses and discussed the need for a critical approach to ensure the validity and accuracy of results. [24]
Mateu has conducted studies on the spatial and spatio-temporal point processes. He conducted research to analyse spatial point patterns across different experimental groups, summarising his findings using the K-function in a non-parametric approach to emphasise the strengths and limitations of spatial data. [25] His work on Functional Data Analysis demonstrated its connection with three traditional types of spatial data structures and provided examples to illustrate the integration of geostatistical data, and areal data. [26] He also introduced a methodological framework based on geostatistics that applied to agricultural planning and environmental restoration. [27] In collaboration with other colleagues, he analysed real-world soil penetration and presented an approach for predicting spatial patterns in functional data which enabled the estimation of values at unobserved locations. [28]
Mateu's research on functional environmental data, particularly in modelling air pollutant concentrations, emphasised the importance of cross-validation for parameter selection and provided insights into adapting kriging techniques. [29] In 2003, he introduced a spatiotemporal Hawkes-type point process model for analysing violence by incorporating daily and weekly periodic patterns in crime occurrences to shed light on the interplay of temporal trends in crime. [30] Expanding on this research, he introduced a deep learning approach in temporal correlations of historical data resulting in the enhancement of police resources, surveillance, crime event predictions, and prevention strategies. [14]