Abstract:
|
Dataset observed over space and time are becoming increasingly abundant and researchers use methods ranging from spatially correlated time series to functional data analysis. There exist a host of methods for modelling using spatially correlated functional data analysis techniques, most of which were motivated by the specific application areas. However, there does not exist any clear pathway to determining the best methodology for a new application. This motives us to present a flow chart of the expected process that one should follow when analysing space-time data. We will demonstrate the use of this framework using a case study, which models the vegetation index of Harvard forest.
|