Functional linear models for the analysis of similarity of waveforms

In seismology methods based on waveform similarity analysis are adopted to identify sequences of events characterized by similar fault mechanism and prop- agation pattern. Seismic waves can be considered as spatially interdependent three dimensional curves depending on time and the waveform similarity analysis can be configured as a functional clustering approach, on the basis of which the member- ship is assessed by the shape of the temporal patterns. For providing qualitative ex- traction of the most important information from the recorded signals we propose an integration of the metadata, related to the waves, as explicative variables of a func- tional linear models. The temporal patterns of this effects, as well as the residual component, are investigated in order to detect a cluster structure. The implemented clustering techniques are based on functional data depth.