A1162
Title: Functional marked ETAS models for the 2009 L'Aquila earthquake sequence
Authors: Giada Adelfio - Universita degli Studi di Palermo (Italy) [presenting]
Abstract: The Epidemic Type Aftershock Sequence (ETAS) model is extended by incorporating functional marks derived from waveform data into the triggering component. Each waveform associated with an event is represented through its functional principal component analysis (FPCA) scores, which are then included as covariates in the triggering term. The model is estimated using the forward likelihood for prediction (FLP) approach and applied to the 2009 L'Aquila earthquake sequence. A local random labelling test for functional marks reveals spatial dependence in the waveform-related features, particularly during the first day of activity. Comparative diagnostics between a standard ETAS model and the proposed functional ETAS model indicate that including waveform information improves the overall goodness-of-fit and the description of the early aftershock clustering.
