George Christakos, Hwa-Lung Yu

Science-Based Spatiotemporal Statistics

Practical Guide with Environmental and Human Exposure Applications

• Gebundenes Buch

Produktbeschreibung

This book provides an introduction to the theoretical development and practical methodology of the so-called science-based spatiotemporal statistics. The book capitalizes on the significance of integrating different knowledge sources (physical, ecological, health, and social) into formal spatiotemporal statistics and provides an array of practical procedures for incorporating these sources into composite space-time analysis, modeling, and estimation/prediction.

Produktdetails

• Verlag: Taylor & Francis Inc
• Seitenzahl: 350
• Erscheinungstermin: 17. August 2028
• Englisch
• Abmessung: 234mm x 156mm
• ISBN-13: 9781482238037
• ISBN-10: 1482238039
• Artikelnr.: 60010918

Autorenporträt

George Christakos is a distinguished professor and S.M. Birch Endowed Chair of the Department of Geography at San Diego State University, California, USA. He is also a Yongqian Chair Professor of the College of Environmental and Resource Sciences at Zhejiang University, China. Hwa-Lung Yu is an associate professor of the Department of Bioenvironmental Systems Engineering at National Taiwan University, Taipei.

Inhaltsangabe

Introduction. A review of spatiotemporal statistics. Science-based vs.
data-processing statistics. Space-time metrics: Euclidean (spatially
isotropic and non-isotropic). Non-Euclidean (river network, gravity model,
isomap). Spatiotemporal random field theory (S/TRF): Ordinary S/TRF.
Generalized S/TRF. Spatiotemporal statistics of data sets. Standard
two-point statistics: Ordinary covariance and variogram. Generalized
covariance and variogram. Extension to multi-point statistics.
Spatiotemporal exploratory analysis (pattern recognition). Empirical
orthogonal method: Atmospheric and hydrological applications. Dynamic
factor analysis: Ecological and environmental applications. Spatiotemporal
clustering analysis: SaTscan and other works. Spatiotemporal statistics of
differential equations. The analytical approach: Closed form expressions of
physical and biophysical laws. Low- and high-order (diagrammatic)
approximations of groundwater flow. The numerical approach: First order
second moment (FOSM) method. Adjoint method. Spatiotemporal trend analysis.
Non-parametric approaches. The generalized additive model. The Kernel
smoothing method. Spatiotemporal regression.