BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260509T055305EDT-71112sUELK@132.216.98.100
DTSTAMP:20260509T095305Z
DESCRIPTION:Title: Exceedance-based nonlinear regression of tail dependence
 .\n\nAbstract: The probability and structure of co-occurrences of extreme 
 values in multivariate data may critically depend on auxiliary information
  provided by covariates. In this talk\, I will develop a flexible generali
 zed additive modelling framework based on high threshold exceedances for e
 stimating covariate-dependent joint tail characteristics for regimes of as
 ymptotic dependence and asymptotic independence. The framework is based on
  suitably defined marginal pretransformations and projections of the rando
 m vector along the directions of the unit simplex\, which lead to convenie
 nt univariate representations of multivariate exceedances based on the exp
 onential distribution. We illustrate this modelling framework on a large d
 ataset of nitrogen dioxide measurements recorded in France between 1999 an
 d 2012\, where we use the generalized additive framework for modelling mar
 ginal distributions and tail dependence in monthly maxima. Results imply a
 symptotic independence of data observed at different stations. We find tha
 t the estimated coefficients of tail dependence decrease as a function of 
 spatial distance. Differences further arise in the patterns for different 
 years and for different types of stations (traffic vs. background).\n
DTSTART:20190403T193000Z
DTEND:20190403T203000Z
LOCATION:Room D4-2019\, CA\, Université de Sherbrooke
SUMMARY:Linda Mhalla\, École des HÉC
URL:https://www.mcgill.ca/mathstat/channels/event/linda-mhalla-ecole-des-he
 c-295786
END:VEVENT
END:VCALENDAR