EcoSta 2023: Start Registration
View Submission - EcoSta2023
A0870
Title: Bayesian spatio-temporal clustering of functional data Authors:  Tomoya Wakayama - The University of Tokyo (Japan)
Genya Kobayashi - Meiji University (Japan)
Shonosuke Sugasawa - Keio University (Japan) [presenting]
Abstract: The use of Bayesian nonparametric modelling and clustering for spatio-temporal functional data is explored. The approach extends a random partition distribution to include spatial similarity, designed explicitly for spatio-temporal scenarios. Efficient algorithms have also been developed to generate posterior samples for this model, making it scalable for large datasets. Simulation studies were conducted and applied to real population data from Tokyo to demonstrate the effectiveness of the method.