Earthquakes can be seen as realization of a spatial, temporal, or spatiotemporal point process. Given a dataset of earthquakes in a mixed geographical region, a scientific question that naturally arises is whether can we separate the earthquakes in two fundamental disjoint sets: triggered (sequential) and background (complete random). Such a separation becomes quite important as background earthquakes are basically blurring main spots of triggered ones. We consider LISA functions as functional marks attached to the points in the spatial point pattern of the earthquakes. We then classify the points through Aitchison distance and subsequent multivariate classification techniques. The performance of our method is demonstrated by simulation.
FUNCTIONAL DATA ANALYSIS FOR SPATIAL AGGREGATED POINT PATTERNS IN SEISMIC SCIENCE
Elvira Romano
;
2019
Abstract
Earthquakes can be seen as realization of a spatial, temporal, or spatiotemporal point process. Given a dataset of earthquakes in a mixed geographical region, a scientific question that naturally arises is whether can we separate the earthquakes in two fundamental disjoint sets: triggered (sequential) and background (complete random). Such a separation becomes quite important as background earthquakes are basically blurring main spots of triggered ones. We consider LISA functions as functional marks attached to the points in the spatial point pattern of the earthquakes. We then classify the points through Aitchison distance and subsequent multivariate classification techniques. The performance of our method is demonstrated by simulation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
