Spatio-Temporal Random Partition Models
share
The number of scientific fields that regularly collect data that are spatio-temporal continues to grow. An intuitive feature of this type of data is that measurements taken on experimental units near each other in time and space tend to be similar. As such, many methods developed to accommodate spatio-temporal dependent structures attempt to borrow strength among units close in space and time, which constitutes an implicit space-time grouping. We develop a class of dependent random partition models that explicitly models this spatio-temporal clustering by way of a dependent random partition model. We first detail how temporal dependence is incorporated so that partitions evolve gently over time. Then conditional and marginal properties of the joint model are derived. We then demonstrate how space can be integrated. Computation strategies are detailed and we illustrate the methodology through simulations and an application.
READ FULL TEXT
