B0242
Title: Modeling multi-way functional data with weak separability
Authors: Kehui Chen - University of Pittsburgh (United States) [presenting]
Abstract: Multi-way functional data refers to an extension where double or multiple indices are involved, such as a sample of brain-imaging data with spatial and temporal indices. In practice, the number of spatial grids and the number of time grids both could be very large, and a multiplication of these two dimensions easily goes beyond the capacity of most data analysis tools. To achieve efficient dimension reduction, one usually adopts the separability assumption that the covariance can be factorized as a product of a spatial covariance and a temporal covariance. We will introduce a new concept of weak separability, and discuss several open questions in factorization methods using the notion of weak separability.
Journal