Estimating the Parameters of a Bivariate Extreme Value Gumbel Distribution

In this paper we compare the estimates based on moments and maximum likelihood criteria with the ones obtained according to the minimum integrated square error criterion. Considering the bivariate extreme value Gumbel distribution, we measure the difference among the parameters estimates resorting to a similarity index between densities, for which a Monte Carlo significance test is introduced. Theory is outlined and main results of a simulation study are provided and commented.