Is the Cox-Stuart test for trend really insensitive to autocorrelation?

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00020699%3A_____%2F14%3A%230000790" target="_blank" >RIV/00020699:_____/14:#0000790 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Is the Cox-Stuart test for trend really insensitive to autocorrelation?

Popis výsledku v původním jazyce

This is an extended abstract related to the poster presentation exhibited during the 7th Global FRIEND-Water conference held in Montpellier, France. The abstract itself was published in the electronic form only as a part of the abstract proceedings recorded on the USB flash drive. The need of its publication arose from the misleading statements in some scientific hydrological articles that recommend the Cox-Stuart (CS) test as being insensitive to the serial correlation when detecting trends. However, by means of Monte Carlo simulations, it was shown that the opposite is true and that the features of this test are even worse than those of the Mann-Kendall test. Therefore, the CS test for trend should be no longer recommended when the autocorrelation ispresent in the series investigated.

Název v anglickém jazyce

Is the Cox-Stuart test for trend really insensitive to autocorrelation?

Popis výsledku anglicky

This is an extended abstract related to the poster presentation exhibited during the 7th Global FRIEND-Water conference held in Montpellier, France. The abstract itself was published in the electronic form only as a part of the abstract proceedings recorded on the USB flash drive. The need of its publication arose from the misleading statements in some scientific hydrological articles that recommend the Cox-Stuart (CS) test as being insensitive to the serial correlation when detecting trends. However, by means of Monte Carlo simulations, it was shown that the opposite is true and that the features of this test are even worse than those of the Mann-Kendall test. Therefore, the CS test for trend should be no longer recommended when the autocorrelation ispresent in the series investigated.

Druh

O - Ostatní výsledky

CEP obor

BB - Aplikovaná statistika, operační výzkum

OECD FORD obor

—

Projekt

—

Návaznosti

V - Vyzkumna aktivita podporovana z jinych verejnych zdroju

Rok uplatnění

2014

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů