In this paper, we give an overview of selected statistical models to draw (statistically significant) conclusions from data. This is particularly important to prove theoretical/physical models. For this reason, we initially describe classical modeling approaches in geostatistics and spatial econometrics. Moreover, we show how large spatial and spatiotemporal data can be modeled by these approaches. Therefore, we sketch selected suitable methods and their computational implementation. Thus, the paper should give a general guideline to analyze big geospatial data, where “big data” refers to the total number of spatially dependent observations, and to draw conclusions from these data. Finally, we compare two of these approaches by applying them to an empirical data set. To be precise, we model the particulate matter concentration (PM2.5) in Colorado.
Bitte loggen Sie sich ein, um Zugang zu Ihrer Lizenz zu erhalten.