Space-time models, methods, and applications

点击翻译按钮获取中文摘要

Many phenomena in nature are measured across space and through time. Investigating possible space-time interactions, in the presence of uncertainty, is key to understanding the science. There is no opportunity to study these interactions when marginalized over time or space. Space-time statistical methodologies should be faithful to methodology in time series analysis and spatial statistics. The models the investigator studies are defined as spatially-dependent filterings of space-time innovation processes that are realizations of geostatistical processes. The innovations do not need to be invariant with respect to time. These models include the usual class of linear time series models, as well as standard geostatistical models, both Gaussian and non-Gaussian. The processes do not need to be stationary in time or in space, and build on the growing literature concerning nonstationary models. Leveraging both the innovations, as well as the filtering operations used to define these space-time processes, this research will springboard from spatial statistics and time series to develop methods for inference and prediction in a space-time context.Ice core paleoclimatology involves the study of the physical and chemical properties preserved in the Earth's glaciers and ice sheets. These properties, analyzed from ice cores, are used as proxies for various climatological processes. In collaboration with other researchers at The Ohio State University, the investigator studies proxy measures of accumulation, extracted from chemical analyses of ice cores distributed over and around Greenland. Using the models developed under this research the investigator will examine relationships between these accumulation records and known drivers of climate variability and change. In addition to climate research, the results will be immediately applicable to other scientific areas, such as speech and hearing sciences. Findings will be communicated via articles in statistical and subject-matter areas. Education of a diverse cross-section of students (statistical and non-statistical), in spatial and time series methodologies, will be achieved in carrying out this research.