In many applications data are collected during their time course where it can no longer be assumed that todays observations are independent from yesterdays. Because these dependencies have to be taken into account for any meaningful statistical analysis, the field of time series analysis aims at investigating, modelling and mathematically analysing them. This is particularly challenging in nonparametric i.e. model-free statistics, where we introduce likelihood approximations for use in bootstrapping and Bayesian analysis - both methods from computational statistics which aim at quantifying uncertainty. We illustrate these approximations using LIGO gravitational wave data.
Claudia Kirch holds a professorship of mathematics at the Ottovon-Guericke university in Magdeburg since 2015 after positions at the University of Kaiserslautern and the Karlsruhe Institute of Technology. She obtained an M. Sc. in mathematics from the Philipps- University of Marburg in 2003 and a PhD from the University of Cologne in 2006. Her research interests focus around time series analysis and nonparametric statistics, where she has worked on mathematical theory, scientific method development as well as applications in neuroscience, remote sensing and finance.
Moderation: Professor Dr. Martin Wendler