Deutsch
 
Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

DATENSATZ AKTIONENEXPORT
Zur Ablage hinzufügen
  Lokale TagsFreigabegeschichteDetailsÜbersicht

Freigegeben
Konferenzbeitrag

Exploring non-parametric tests for power evaluation in trend detection

Urheber*innen

Lioi,  Beatrice
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Totaro,  Vincenzo
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Kochanek,  Krzysztof
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Gioia,  Andrea
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Bisantino,  Tiziana
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Iacobellis,  Vito
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Externe Ressourcen
Es sind keine externen Ressourcen hinterlegt
Volltexte (frei zugänglich)
Es sind keine frei zugänglichen Volltexte in GFZpublic verfügbar
Ergänzendes Material (frei zugänglich)
Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar
Zitation

Lioi, B., Totaro, V., Kochanek, K., Gioia, A., Bisantino, T., Iacobellis, V. (2023): Exploring non-parametric tests for power evaluation in trend detection, XXVIII General Assembly of the International Union of Geodesy and Geophysics (IUGG) (Berlin 2023).
https://doi.org/10.57757/IUGG23-2019


Zitierlink: https://gfzpublic.gfz-potsdam.de/pubman/item/item_5018835
Zusammenfassung
As a consequence of the growing concerns about the risks caused by climate change in the frequency of extreme events, the community of scientists tackled this topic from different perspectives, among which there is led to further investigate the climate change. One of the research for deterministic changes affecting stochastic processes still needs further confirmation in observations to provide more valid and robust results. Several approached are available in the current scientific literature for the detection of trend in time series, whose nature can be traceable to the dualism between parametric and non-parametric statistical testing. However, the use of non-parametric tests has been so far less exploited. In any case, has to be stressed that the use of any of these tests should undergo an evaluation of their power of the test, strictly correlated to the existence of the Type II Error, whose neglection is susceptible to alter the preparedness level in the framework of infrastructure design. An increasing number of researches is devoted to the numerical evaluation of the power of both parametric and non-parametric tests, assuming a threshold value for considering the use of test result acceptable. However, these studies typically rely on the assumption of a linear deterministic trend on the parent distribution mean value. The question is whether more complex non-stationary assumptions may further affect the test power? This work is aimed at answering this question, which could also help stakeholders in more efficient design and management of civil infrastructures in a changing climatic context.