The Bernstein Center Freiburg




Informal Seminar
Marcel Sauerbier

Faculty of Engineering
University of Freiburg


Efficient Cumulant Estimation

Cumulants are measures related to higher-order correlations. As their estimation through classical k-statistics is computationally expensive and statistically complex, a new method to simplify cumulant estimation for Hawkes processes was recently proposed by Stojan Jovanovic, Stefan Rotter and John Hertz. In my talk I will present the basic idea behind this simplified method as well as its performance in comparison with k-statistics. A simulation program for uni- and multivariate Hawkes processes and a program for cumulant analysis are thus introduced.

Friday, November 28, 2014

14:00 h
BCF Library
First Floor
Hansastr. 9a
The talk is open to the public. Guests are cordially invited!
www.bcf.uni-freiburg.de