New study reveals key to predicting extreme events in multivariate data.
The article explores how to model heavy-tailed phenomena using a mathematical tool called multivariate regular variation (MRV). By studying the tail dependence function of the copula associated with these distributions, the researchers found that the lower tail dependence function of the survival copula is crucial for a random vector with heavy tails to have a MRV tail. They also identified the limit measure of the MRV tail and extended their analysis to other heavy-tailed distribution families like subexponential and long-tailed distributions.