A1081
Title: Motif discovery driven forecasting for functional data
Authors: Jacopo Di Iorio - Penn State University (United States) [presenting]
Abstract: Forecasting has always been a major goal of functional data analysis, involving the prediction of future values and/or the evolution of functional observations. Given the increasing attention in the field of functional motif discovery, functional forecasting is performed through the identification of functional motifs. Functional motifs represent typical "shapes" or "patterns" recurring multiple times within a single curve and/or across misaligned portions of multiple curves. Portions characterized by the same motif are hypothesized to be more likely to evolve similarly. Extensive diagnostics can guide the user not only in tuning parameters but also in validating the aforementioned hypothesis, thus ensuring the applicability of the method. Method performance is assessed through simulations and is applied to a real-data case study.
