A1563
Title: Variograms for Kriging and clustering of spatial functional data with phase variation
Authors: Sebastian Kurtek - The Ohio State University (United States) [presenting]
Karthik Bharath - University of Nottingham (United Kingdom)
Xiaohan Guo - Pfizer (United States)
Abstract: Spatial, amplitude and phase variations in spatial functional data are confounded. Conclusions from the popular functional trace-variogram, which quantifies spatial variation, can be misleading when analyzing misaligned functional data with phase variation. To remedy this, a framework is described that extends amplitude-phase separation methods in functional data to the spatial setting, with a view towards performing clustering and spatial prediction. A decomposition of the trace-variogram is proposed into amplitude and phase components and quantifies how spatial correlations between functional observations manifest in their respective amplitude and phase. This enables the generation of separate amplitude and phase clustering methods for spatial functional data and develops a novel spatial functional interpolant at unobserved locations based on combining separate amplitude and phase predictions. Through simulations and real data analyses, we demonstrate the advantages of the approach when compared to standard ones that ignore phase variation through more accurate predictions and more interpretable clustering results.
