Graduate Seminars

back Back to all Graduate Seminars

AMCS Graduate Seminar| Bridging Asymptotic Independence and Dependence in Spatial Extremes Using Gaussian Scale Mixtures

Start Date: September 14, 2017
End Date: September 14, 2017

By Professor Raphaël Huser (KAUST)
Gaussian scale mixtures are constructed as Gaussian processes with a random variance. They have non-Gaussian marginals and can exhibit asymptotic dependence unlike Gaussian processes, which are asymptotically independent except in the case of perfect dependence. In this graduate seminar, I will explain what Gaussian scale mixtures are, and why they are useful for the modeling of spatial extreme events. Focusing on one particular model able to transition between asymptotic dependence classes, I will then show how inference may be performed based on high threshold exceedances using a censored likelihood approach. This methodology will then be illustrated with wind speed data collected in the Pacific Northwest, US, and the model will be shown to adequately capture the data's extremal properties.
Biography: Raphaël Huser is an Assistant Professor of Statistics in the Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division at KAUST. He completed his PhD in 2013 at the Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland, and spent about a year as a Postdoctoral Research Fellow at KAUST in 2014. He received the EPFL Doctorate Award 2014 to his PhD Thesis, the Lambert Award 2015 from the Swiss Statistical Society, and got Elected to the International Statistical Institute (ISI) in 2016. He is currently serving as an Associate Editor for the journal Extremes. Raphaël Huser's main research interests lie at the intersection between statistics of extreme events, risk assessment, spatio-temporal statistics, dependence modeling for complex systems, and statistical approaches for large datasets, with particular focus on environmental applications such as the prediction of extreme flooding, droughts, and wind gusts. He develops statistical models for rare events, as well as efficient inference methods to fit them to data.

More Information:

For more info contact: Prof. Raphael G. Huser : email:
Date: Thursday 14th Sep 2017
Time:12:00 PM - 01:00 PM
Location: Building 9, Lecture Hall 1 Room 2322
Refreshments: Light Lunch will be served at 11:45