Articles | Volume 24, issue 1
https://doi.org/10.5194/hess-24-473-2020
https://doi.org/10.5194/hess-24-473-2020
Research article
 | 
29 Jan 2020
Research article |  | 29 Jan 2020

Numerical investigation on the power of parametric and nonparametric tests for trend detection in annual maximum series

Vincenzo Totaro, Andrea Gioia, and Vito Iacobellis

Viewed

Total article views: 3,904 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
2,828 1,005 71 3,904 78 69
  • HTML: 2,828
  • PDF: 1,005
  • XML: 71
  • Total: 3,904
  • BibTeX: 78
  • EndNote: 69
Views and downloads (calculated since 08 Aug 2019)
Cumulative views and downloads (calculated since 08 Aug 2019)

Viewed (geographical distribution)

Total article views: 3,904 (including HTML, PDF, and XML) Thereof 3,198 with geography defined and 706 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 23 Nov 2024
Download
Short summary
We highlight the need for power evaluation in the application of null hypothesis significance tests for trend detection in extreme event analysis. In a wide range of conditions, depending on the underlying distribution of data, the test power may reach unacceptably low values. We propose the use of a parametric approach, based on model selection criteria, that allows one to choose the null hypothesis, to select the level of significance, and to check the test power using Monte Carlo experiments.