This project will derive space-time explicit models and measurements that can parameterize spatially inter-influenced patterns among space, time, and ecological covariates. By defining parameterized spatial associations in the context of generalized linear models and generalized linear mixed-effect models, this project will provide a package of spatial measurements to quantify the shape, size, and value gradient of spatially clustered data that may change over time. Since it is unlikely to have a closed form for all the estimation methods, the project will investigate frequentist, Bayesian, and numerical methods for statistical estimations and parameterizations of spatially clustered data.
Results from the project are likely to have a number of impacts. It generally is accepted that a wide range of spatial factors influence spatial disparities, spatial grouping, and reorganization of various natural and human environments. Statistical studies of spatial patterning are able to account for these issues while revealing various spatial associations and correlations. As a result, there is a growing interest in spatial methods in fields such as economics, criminology, demography, and population health. For example, a quick parameterized surveillance of spatial patterns and signals and their strength on the basis of known risk factors will add the nation's quick response capabilities for potential harms. Moreover, the training of graduate students during the project years and beyond will foster interdisciplinary learning and groom the next generation of scholars who will help develop quantitative methods that can benefit academic disciplines and society as a whole.
