Bayesian nonparametric estimation of the intensities in multivariate Hawkes processes

Add to your list(s) Download to your calendar using vCal

• Judith Rousseau (Oxford)
• Friday 26 January 2018, 16:00-17:00
• MR12.

If you have a question about this talk, please contact Quentin Berthet.

Joint wortk with Sophie Donnet and Vincent Rivoirard.

Hawkes processes are special cases of point processes. In some cases they are also called self excited Poisson processes. Generally speaking, if (N_t, t ∈ [0, T])is the point process, Hawkes processes are caracterised by having a conditional intensity function given by λ(t) = (ν +int_0^t− h(t − u)dN_u)_+.

In this work we are interested in estimating the parameters (ν, h) in the context of multivariate Hawkes processes, using Bayesian nonparametric approaches. We propose generic conditions on the true parameters ν, h and the associated prior distributions to obtain posterior concentration rates under the L1 norm for these parameters. We apply these conditions to various families of prior models and finally we present a simulation study in the context of neuroscience.

This talk is part of the Statistics series.

This talk is included in these lists:

• All CMS events
• All Talks (aka the CURE list)
• CMS Events
• Cambridge Forum of Science and Humanities
• Cambridge Language Sciences
• Cambridge talks
• Chris Davis' list
• DPMMS Lists
• DPMMS info aggregator
• DPMMS lists
• Guy Emerson's list
• Hanchen DaDaDash
• Interested Talks
• MR12
• Machine Learning
• School of Physical Sciences
• Statistical Laboratory info aggregator
• Statistics
• Statistics Group
• bld31
• custom
• rp587

Note that ex-directory lists are not shown.