Result Details

Automated threshold selection for parametric and non-parametric estimates of intensity-duration-frequency curves

HOLEŠOVSKÝ, J.; FUSEK, M.; MICHÁLEK, J. Automated threshold selection for parametric and non-parametric estimates of intensity-duration-frequency curves. In Mathematical Methods in Science and Engineering. Advances in Mathematical and Computational Methods. Mathematics and Computers in Science and Engineering Series. Athens: 2014. p. 114-118. ISBN: 978-1-61804-256-9. ISSN: 2227-4588.
Type
conference paper
Language
English
Authors
Holešovský Jan, Ing., Ph.D., IM (FME)
Fusek Michal, Ing., Ph.D., UMAT (FEEC)
Michálek Jaroslav, doc. RNDr., CSc., FME (FME), IM (FME)
Abstract

When dealing with extreme values estimation, the threshold models are often used. However, a proper threshold selection is one of the problems which have to be solved. In this paper, we concentrate on this issue in order to compare an automated threshold selection based on multiple-threshold model with double bootstrap technique based on semi-parametric estimation. A case study is carried out to evaluate estimations of intensity-duration-frequency curves which represent commonly used hydrological tool. A special attention is also paid to the assessment of an impact of the series length on the estimation quality.

Keywords

extreme value theory; threshold selection; IDF curves; bootstrap

Published
2014
Pages
114–118
Journal
Advances in Mathematical and Computational Methods, ISSN 2227-4588
Proceedings
Mathematical Methods in Science and Engineering
Series
Mathematics and Computers in Science and Engineering Series
Conference
Mathematical Methods & Computational Techniques in Science & Engineering
ISBN
978-1-61804-256-9
Place
Athens
BibTeX
@inproceedings{BUT110889,
  author="Jan {Holešovský} and Michal {Fusek} and Jaroslav {Michálek}",
  title="Automated threshold selection for parametric and non-parametric estimates of intensity-duration-frequency curves",
  booktitle="Mathematical Methods in Science and Engineering",
  year="2014",
  series="Mathematics and Computers in Science and Engineering Series",
  journal="Advances in Mathematical and Computational Methods",
  pages="114--118",
  address="Athens",
  isbn="978-1-61804-256-9",
  issn="2227-4588"
}
Departments
Back to top