Séminaire

Intervening organization
AgroParisTech-INRA, UMR 518 MIA
Name of intervener
Sophie DONNET
Title
Bayesian estimation for multidimensional Hawkes processes
Abstract

Multidimensional Hawkes processes are used to modelise multivariate neuron spike data. The estimation of intensity functions allows to understand the neuronal interaction structure. In a non-parametric frequentist framework, LASSO estimators have been proposed in the literature. In this work, we propose a Bayesian non-parametric estimation. We sample the posterior distribution through a Sequential Monte Carlo algorithm, well adapted to point processes.

Place
Salle de réunion 142, bâtiment 210
Date of the day