Moment estimation for multivariate extreme value distribution is describedin this paper. Asymptotic covariance matrix of the estimators is given. The relativeefficiencies of moment estimators as compared with the maxi...Moment estimation for multivariate extreme value distribution is describedin this paper. Asymptotic covariance matrix of the estimators is given. The relativeefficiencies of moment estimators as compared with the maximum likelihood and thestepwise estimators are computed. We show that when there is strong dependencebetween the variates, the generalized variance of moment estimators is much lower thanthe stepwise estimators. It becomes more obvious when the dimension increases.展开更多
文摘Moment estimation for multivariate extreme value distribution is describedin this paper. Asymptotic covariance matrix of the estimators is given. The relativeefficiencies of moment estimators as compared with the maximum likelihood and thestepwise estimators are computed. We show that when there is strong dependencebetween the variates, the generalized variance of moment estimators is much lower thanthe stepwise estimators. It becomes more obvious when the dimension increases.
