Nonparametric analysis of the time structure of seismicity in a geographic region

As an alternative to traditional parametric approaches, we suggest nonparametric methods for analyzing temporal data on earthquake occurrences. In particular, the kernel method for estimating the hazard function and the intensity function are presented. One novelty of our approaches is that we take into account the possible dependence of the data to estimate the distribution of time intervals between earthquakes, which has not been considered in most statistics studies on seismicity. Kernel estimation of hazard function has been used to study the occurrence process of cluster centers (main shocks). Kernel intensity estimation, on the other hand, has helped to describe the occurrence process of cluster members (aftershocks). Similar studies in two geographic areas of Spain (Granada and Galicia) have been carried out to illustrate the estimation methods suggested.