Two new methods are suggested for estimating the dependence function of a bivariate extreme-value distribution. One is based on a multiplicative modification of an earlier technique proposed by Pickands, and the other employs spline smoothing under constraints. Both produce estimators that satisfy all the conditions that define a dependence function, including convexity and the restriction that its curve lie within a certain triangular region. The first approach does not require selection of smoothing parameters; the second does, and for that purpose we suggest explicit tuning methods, one of them based on cross-validation.
@article{1081282691,
author = {Hall, Peter and Tajvidi, Nader},
title = {Distribution and dependence-function estimation for bivariate extreme-value distributions},
journal = {Bernoulli},
volume = {6},
number = {6},
year = {2000},
pages = { 835-844},
language = {en},
url = {http://dml.mathdoc.fr/item/1081282691}
}
Hall, Peter; Tajvidi, Nader. Distribution and dependence-function estimation for bivariate extreme-value distributions. Bernoulli, Tome 6 (2000) no. 6, pp. 835-844. http://gdmltest.u-ga.fr/item/1081282691/