作者: Stephan Schlüter , Matthias Fischer
DOI: 10.1080/03610926.2010.513784
关键词:
摘要: In this article, we derive a new formula for extreme Student t quantiles. We use the fact that distribution arises as limit of variance-mixture normals. For normal there is already tail quantile derived by Reiss (1989). generalize his procedure and transfer it to our scenario. Eventually, compare estimates those from Gafer Kafadar (1984), who also formula. Using R generate benchmark find method more accurate very high