A new method to simulate Hawkes processes
In Neuroscience, Hawkes process is one of the most popular model to capture the activity of neural networks. In this talk, we are interested in simulating Hawkes processes with a huge number of neurons. By using Kalikow decomposition, we modify the classical Ogata algorithm to obtain a new algorithm, that is more tractable in practice. We present some interesting mathematical and simulated results. This based on recent works with Patricia Reynaud-Bouret (LJAD), Alexandre Muzy (I3S, Inria) and Eva Locherbach (Paris 1). This is a joint work with Paul Gresland (NeuroMod).