Angulo, José MiguelJosé MiguelAnguloHWA-LUNG YULangousis, AndreasAndreasLangousisMadrid, Ana EstherAna EstherMadridChristakos, GeorgeGeorgeChristakos2018-09-102018-09-102012http://www.scopus.com/inward/record.url?eid=2-s2.0-84867244309&partnerID=MN8TOARShttp://scholars.lib.ntu.edu.tw/handle/123456789/373018A theoretical model for the spread of infectious diseases in a composite space-time domain is developed. The model has a general form that enables it to account for the basic mechanisms of disease distribution and to incorporate the considerable multisourced uncertainty (caused by physiographic features, disease variability, meteorological conditions, etc.). Starting from the general model formulation regarding the specification of transmission and recovery rates, as well as the population migration dynamics, several subsequent assumptions are introduced that simplify analytical tractability and practical implementation. In particular, linearization involving a deterministic functional representation for the average evolution of the fraction of susceptible individuals allows the formulation of an extended Kalman filter approach for estimation based on the time series observed at a finite set of locations. Different aspects of interest derived from the epidemic space-time model proposed, as well as the performance of the extended Kalman filter procedure, are illustrated through simulations. ? 2012 Copyright Taylor and Francis Group, LLC.geostatistics; spatiotemporal data modeling; uncertainty[SDGs]SDG3computer simulation; disease spread; geostatistics; infectious disease; Kalman filter; numerical model; population migration; spatiotemporal analysis; time series analysis; uncertainty analysisjournal article10.1080/13658816.2011.648642