作者: Margarida Brito , Laura Cavalcante , Ana Cristina Moreira Freitas
DOI: 10.1088/1751-8113/49/21/214003
关键词: Extreme value theory 、 Applied mathematics 、 Estimator 、 Asymptotic theory (statistics) 、 Statistics 、 Mathematics 、 Order (group theory) 、 Confidence interval 、 Type (model theory) 、 Tail index
摘要: The estimation of the tail index is a central topic in extreme value analysis. We consider geometric-type estimator for and study its asymptotic properties. propose here two equivalent bias-corrected estimators establish corresponding behaviour. also apply suggested to construct confidence intervals this parameter. Some simulations order illustrate finite sample behaviour proposed are provided.