Last week, international researchers met to discuss recent mathematical developments in risk modeling. 52°North’s Dr. Ben Gräler demonstrated the use of local spatial and spatio-temporal vine copulas to model extreme environmental phenomena.

The Centre de Recherches de Mathématiques recently organized the Workshop “Dependence modeling tools for risk management” at the University of Montreal. More than 50 international researchers attended the workshop to discuss recent mathematical developments in risk modeling from theory to applications in finance, econometrics, insurance and spatial statistics. 52°North’s Dr Benedikt Gräler gave an invited talk on “Modeling Extremes with Local Spatial and Spatio-Temporal Vine Copulas”. He explained how understanding and modeling extreme environmental phenomena often requires non-Gaussian dependence structures and heavily skewed marginal distribution functions. Local spatial and spatio-temporal vine copulas allow for the necessary flexibility.