| Classical multivariate extreme value theory tries to capture the extremal dependence between components under a multivariate domain of attraction condition and it requires each of the components to be in domain of attraction of a univariate extreme value distribution. Multivariate extreme value (MEV) has a rich theory but has some backdraws as it cannot capture the difference between asymptotic independence and asymptotic dependence. A different approach to MEV was given by Heffernan and Tawn (2004) where they examined MEV distributions by conditioning on one of the components to be extreme. In this talk we shall discuss this model in the bivariate situation and see how the product of the components behave under this model. |