With Mathieu Boudreault, we finally uploaded our working paper on multivariate integer-valued autoregressive models applied to earthquake counts on http://hal.archives-ouvertes.fr/ and on http://arxiv.org/.
"In various situations in the insurance industry, in finance, in epidemiology, etc., one needs to represent the joint evolution of the number of occurrences of an event. In this paper, we present a multivariate integer-valued autoregressive (MINAR) model, derive its properties and apply the model to earthquake occurrences across various pairs of tectonic plates. The model is an extension of Pedelis & Karlis (2011) where cross autocorrelation (spatial contagion in a seismic context) is considered. We fit various bivariate count models and find that for many contiguous tectonic plates, spatial contagion is significant in both directions. Furthermore, ignoring cross autocorrelation can underestimate the potential for high numbers of occurrences over the short-term. Our overall findings seem to further confirm Parsons & Velasco (2001)."
The starting point of our paper with Mathieu was the paper on the absence of remotely triggered large earthquakes beyond the main shock region, by Thomas Parsons and Aaron Velasco published in May 2011 in Nature Geoscience. I was supposed to present this work at the Geotop seminar last week, but the seminar has been canceled and I will probably present it this Winter. Slides as well as R code will be uploaded for the seminar.