Our very best wishes for the New Year!
The novelty of this year is that our seminars will now take place simultaneously at ESSEC Paris La Défense (CNIT) and at ESSEC Asia Pacific (5 Nepal Park, Singapore). Our new Essec colleague in Statistics (Singapore campus), Prof. Jeremy Heng, has joined the organizing team for the WG Risk seminars series; welcome, Jeremy!
This quarter, we will go on focusing on data science or/and cyber risk & security, with various experts from academic or professional institutions discussing methods/technics of data analytics (or more generally data science) and applications & problems linked to risk analysis and management, with a specific interest on cyber risk & security.
For our 1st meeting of 2019, we have the pleasure, thanks to the support of the ESSEC IDS dpt, Institut des Actuaires, LabEx MME-DII, and the group BFA (SFdS), to invite:
Prof. Justin DAUWELS
School of Electrical and Electronic Engineering
Nanyang Technological University - NTU Singapore
who will speak on
Graphical Models for Extreme Events
Date: Thursday, January 17, 12:30 - 1:30pm (CET)
Venue: ESSEC Paris La Défense (CNIT) - room 202 / ESSEC Asia Pacific - Level 3, classroom 7 (at 7:30pm Singapore time)
Abstract: Assessing the risk of extreme events, such as hurricanes, floods, and droughts, presents unique significance in practice. Unfortunately, the existing extreme-value statistical models are typically not feasible for practical large-scale problems. Graphical models, on the other hand, are capable of handling a sizable number of variables, but have yet to be explored in the realm of extreme-value analysis. To bridge the gap, we present how to utilize graphical models to analyze extreme events in this talk. Extreme events are often modeled in two stages: first the extreme-value marginal distributions are estimated (i.e., marginal analysis), and then the joint distribution of extreme values is constructed based on the marginals (i.e., joint analysis). We construct graphical models for both marginal and joint analysis problems, show theoretical properties, and further apply them to various real extreme-value data sets.
Jeremy Heng, Olga Klopp, Marie Kratz, Isabelle Wattiau